diff options
Diffstat (limited to 'drivers')
63 files changed, 2880 insertions, 1240 deletions
diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index 091aef281963..4d9e2f855e9d 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -83,6 +83,7 @@ i915-y += i915_cmd_parser.o \ i915-y += intel_uc.o \ intel_uc_fw.o \ intel_guc.o \ + intel_guc_ads.o \ intel_guc_ct.o \ intel_guc_fw.o \ intel_guc_log.o \ diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index e968aeae1d84..3849ded354e3 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -988,7 +988,10 @@ i915_next_seqno_set(void *data, u64 val) if (ret) return ret; + intel_runtime_pm_get(dev_priv); ret = i915_gem_set_global_seqno(dev, val); + intel_runtime_pm_put(dev_priv); + mutex_unlock(&dev->struct_mutex); return ret; @@ -2464,24 +2467,11 @@ static int i915_guc_log_control_get(void *data, u64 *val) static int i915_guc_log_control_set(void *data, u64 val) { struct drm_i915_private *dev_priv = data; - int ret; if (!HAS_GUC(dev_priv)) return -ENODEV; - if (!dev_priv->guc.log.vma) - return -EINVAL; - - ret = mutex_lock_interruptible(&dev_priv->drm.struct_mutex); - if (ret) - return ret; - - intel_runtime_pm_get(dev_priv); - ret = i915_guc_log_control(dev_priv, val); - intel_runtime_pm_put(dev_priv); - - mutex_unlock(&dev_priv->drm.struct_mutex); - return ret; + return intel_guc_log_control(&dev_priv->guc, val); } DEFINE_SIMPLE_ATTRIBUTE(i915_guc_log_control_fops, @@ -2518,15 +2508,19 @@ static int i915_edp_psr_status(struct seq_file *m, void *data) u32 stat[3]; enum pipe pipe; bool enabled = false; + bool sink_support; if (!HAS_PSR(dev_priv)) return -ENODEV; + sink_support = dev_priv->psr.sink_support; + seq_printf(m, "Sink_Support: %s\n", yesno(sink_support)); + if (!sink_support) + return 0; + intel_runtime_pm_get(dev_priv); mutex_lock(&dev_priv->psr.lock); - seq_printf(m, "Sink_Support: %s\n", yesno(dev_priv->psr.sink_support)); - seq_printf(m, "Source_OK: %s\n", yesno(dev_priv->psr.source_ok)); seq_printf(m, "Enabled: %s\n", yesno((bool)dev_priv->psr.enabled)); seq_printf(m, "Active: %s\n", yesno(dev_priv->psr.active)); seq_printf(m, "Busy frontbuffer bits: 0x%03x\n", @@ -2584,9 +2578,9 @@ static int i915_edp_psr_status(struct seq_file *m, void *data) seq_printf(m, "Performance_Counter: %u\n", psrperf); } if (dev_priv->psr.psr2_support) { - u32 psr2 = I915_READ(EDP_PSR2_STATUS_CTL); + u32 psr2 = I915_READ(EDP_PSR2_STATUS); - seq_printf(m, "EDP_PSR2_STATUS_CTL: %x [%s]\n", + seq_printf(m, "EDP_PSR2_STATUS: %x [%s]\n", psr2, psr2_live_status(psr2)); } mutex_unlock(&dev_priv->psr.lock); @@ -2710,7 +2704,8 @@ static int i915_runtime_pm_status(struct seq_file *m, void *unused) if (!HAS_RUNTIME_PM(dev_priv)) seq_puts(m, "Runtime power management not supported\n"); - seq_printf(m, "GPU idle: %s\n", yesno(!dev_priv->gt.awake)); + seq_printf(m, "GPU idle: %s (epoch %u)\n", + yesno(!dev_priv->gt.awake), dev_priv->gt.epoch); seq_printf(m, "IRQs disabled: %s\n", yesno(!intel_irqs_enabled(dev_priv))); #ifdef CONFIG_PM @@ -3143,8 +3138,8 @@ static int i915_engine_info(struct seq_file *m, void *unused) intel_runtime_pm_get(dev_priv); - seq_printf(m, "GT awake? %s\n", - yesno(dev_priv->gt.awake)); + seq_printf(m, "GT awake? %s (epoch %u)\n", + yesno(dev_priv->gt.awake), dev_priv->gt.epoch); seq_printf(m, "Global active requests: %d\n", dev_priv->gt.active_requests); seq_printf(m, "CS timestamp frequency: %u kHz\n", @@ -3363,7 +3358,10 @@ static void drrs_status_per_crtc(struct seq_file *m, /* disable_drrs() will make drrs->dp NULL */ if (!drrs->dp) { - seq_puts(m, "Idleness DRRS: Disabled"); + seq_puts(m, "Idleness DRRS: Disabled\n"); + if (dev_priv->psr.enabled) + seq_puts(m, + "\tAs PSR is enabled, DRRS is not enabled\n"); mutex_unlock(&drrs->mutex); return; } @@ -4606,6 +4604,46 @@ static const struct file_operations i915_hpd_storm_ctl_fops = { .write = i915_hpd_storm_ctl_write }; +static int i915_drrs_ctl_set(void *data, u64 val) +{ + struct drm_i915_private *dev_priv = data; + struct drm_device *dev = &dev_priv->drm; + struct intel_crtc *intel_crtc; + struct intel_encoder *encoder; + struct intel_dp *intel_dp; + + if (INTEL_GEN(dev_priv) < 7) + return -ENODEV; + + drm_modeset_lock_all(dev); + for_each_intel_crtc(dev, intel_crtc) { + if (!intel_crtc->base.state->active || + !intel_crtc->config->has_drrs) + continue; + + for_each_encoder_on_crtc(dev, &intel_crtc->base, encoder) { + if (encoder->type != INTEL_OUTPUT_EDP) + continue; + + DRM_DEBUG_DRIVER("Manually %sabling DRRS. %llu\n", + val ? "en" : "dis", val); + + intel_dp = enc_to_intel_dp(&encoder->base); + if (val) + intel_edp_drrs_enable(intel_dp, + intel_crtc->config); + else + intel_edp_drrs_disable(intel_dp, + intel_crtc->config); + } + } + drm_modeset_unlock_all(dev); + + return 0; +} + +DEFINE_SIMPLE_ATTRIBUTE(i915_drrs_ctl_fops, NULL, i915_drrs_ctl_set, "%llu\n"); + static const struct drm_info_list i915_debugfs_list[] = { {"i915_capabilities", i915_capabilities, 0}, {"i915_gem_objects", i915_gem_object_info, 0}, @@ -4683,7 +4721,8 @@ static const struct i915_debugfs_files { {"i915_dp_test_active", &i915_displayport_test_active_fops}, {"i915_guc_log_control", &i915_guc_log_control_fops}, {"i915_hpd_storm_ctl", &i915_hpd_storm_ctl_fops}, - {"i915_ipc_status", &i915_ipc_status_fops} + {"i915_ipc_status", &i915_ipc_status_fops}, + {"i915_drrs_ctl", &i915_drrs_ctl_fops} }; int i915_debugfs_register(struct drm_i915_private *dev_priv) diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index 173d0095e3b2..e9f1daf258fe 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -55,6 +55,7 @@ static struct drm_driver driver; +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) static unsigned int i915_load_fail_count; bool __i915_inject_load_failure(const char *func, int line) @@ -70,6 +71,7 @@ bool __i915_inject_load_failure(const char *func, int line) return false; } +#endif #define FDO_BUG_URL "https://bugs.freedesktop.org/enter_bug.cgi?product=DRI" #define FDO_BUG_MSG "Please file a bug at " FDO_BUG_URL " against DRM/Intel " \ @@ -107,8 +109,12 @@ __i915_printk(struct drm_i915_private *dev_priv, const char *level, static bool i915_error_injected(struct drm_i915_private *dev_priv) { +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) return i915_modparams.inject_load_failure && i915_load_fail_count == i915_modparams.inject_load_failure; +#else + return false; +#endif } #define i915_load_error(dev_priv, fmt, ...) \ @@ -176,96 +182,103 @@ static void intel_detect_pch(struct drm_i915_private *dev_priv) * of only checking the first one. */ while ((pch = pci_get_class(PCI_CLASS_BRIDGE_ISA << 8, pch))) { - if (pch->vendor == PCI_VENDOR_ID_INTEL) { - unsigned short id = pch->device & INTEL_PCH_DEVICE_ID_MASK; - - dev_priv->pch_id = id; - - if (id == INTEL_PCH_IBX_DEVICE_ID_TYPE) { - dev_priv->pch_type = PCH_IBX; - DRM_DEBUG_KMS("Found Ibex Peak PCH\n"); - WARN_ON(!IS_GEN5(dev_priv)); - } else if (id == INTEL_PCH_CPT_DEVICE_ID_TYPE) { - dev_priv->pch_type = PCH_CPT; - DRM_DEBUG_KMS("Found CougarPoint PCH\n"); - WARN_ON(!IS_GEN6(dev_priv) && - !IS_IVYBRIDGE(dev_priv)); - } else if (id == INTEL_PCH_PPT_DEVICE_ID_TYPE) { - /* PantherPoint is CPT compatible */ - dev_priv->pch_type = PCH_CPT; - DRM_DEBUG_KMS("Found PantherPoint PCH\n"); - WARN_ON(!IS_GEN6(dev_priv) && - !IS_IVYBRIDGE(dev_priv)); - } else if (id == INTEL_PCH_LPT_DEVICE_ID_TYPE) { - dev_priv->pch_type = PCH_LPT; - DRM_DEBUG_KMS("Found LynxPoint PCH\n"); - WARN_ON(!IS_HASWELL(dev_priv) && - !IS_BROADWELL(dev_priv)); - WARN_ON(IS_HSW_ULT(dev_priv) || - IS_BDW_ULT(dev_priv)); - } else if (id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) { - dev_priv->pch_type = PCH_LPT; - DRM_DEBUG_KMS("Found LynxPoint LP PCH\n"); - WARN_ON(!IS_HASWELL(dev_priv) && - !IS_BROADWELL(dev_priv)); - WARN_ON(!IS_HSW_ULT(dev_priv) && - !IS_BDW_ULT(dev_priv)); - } else if (id == INTEL_PCH_WPT_DEVICE_ID_TYPE) { - /* WildcatPoint is LPT compatible */ - dev_priv->pch_type = PCH_LPT; - DRM_DEBUG_KMS("Found WildcatPoint PCH\n"); - WARN_ON(!IS_HASWELL(dev_priv) && - !IS_BROADWELL(dev_priv)); - WARN_ON(IS_HSW_ULT(dev_priv) || - IS_BDW_ULT(dev_priv)); - } else if (id == INTEL_PCH_WPT_LP_DEVICE_ID_TYPE) { - /* WildcatPoint is LPT compatible */ - dev_priv->pch_type = PCH_LPT; - DRM_DEBUG_KMS("Found WildcatPoint LP PCH\n"); - WARN_ON(!IS_HASWELL(dev_priv) && - !IS_BROADWELL(dev_priv)); - WARN_ON(!IS_HSW_ULT(dev_priv) && - !IS_BDW_ULT(dev_priv)); - } else if (id == INTEL_PCH_SPT_DEVICE_ID_TYPE) { - dev_priv->pch_type = PCH_SPT; - DRM_DEBUG_KMS("Found SunrisePoint PCH\n"); - WARN_ON(!IS_SKYLAKE(dev_priv) && - !IS_KABYLAKE(dev_priv)); - } else if (id == INTEL_PCH_SPT_LP_DEVICE_ID_TYPE) { - dev_priv->pch_type = PCH_SPT; - DRM_DEBUG_KMS("Found SunrisePoint LP PCH\n"); - WARN_ON(!IS_SKYLAKE(dev_priv) && - !IS_KABYLAKE(dev_priv)); - } else if (id == INTEL_PCH_KBP_DEVICE_ID_TYPE) { - dev_priv->pch_type = PCH_KBP; - DRM_DEBUG_KMS("Found Kaby Lake PCH (KBP)\n"); - WARN_ON(!IS_SKYLAKE(dev_priv) && - !IS_KABYLAKE(dev_priv) && - !IS_COFFEELAKE(dev_priv)); - } else if (id == INTEL_PCH_CNP_DEVICE_ID_TYPE) { - dev_priv->pch_type = PCH_CNP; - DRM_DEBUG_KMS("Found Cannon Lake PCH (CNP)\n"); - WARN_ON(!IS_CANNONLAKE(dev_priv) && - !IS_COFFEELAKE(dev_priv)); - } else if (id == INTEL_PCH_CNP_LP_DEVICE_ID_TYPE) { - dev_priv->pch_type = PCH_CNP; - DRM_DEBUG_KMS("Found Cannon Lake LP PCH (CNP-LP)\n"); - WARN_ON(!IS_CANNONLAKE(dev_priv) && - !IS_COFFEELAKE(dev_priv)); - } else if (id == INTEL_PCH_P2X_DEVICE_ID_TYPE || - id == INTEL_PCH_P3X_DEVICE_ID_TYPE || - (id == INTEL_PCH_QEMU_DEVICE_ID_TYPE && - pch->subsystem_vendor == - PCI_SUBVENDOR_ID_REDHAT_QUMRANET && - pch->subsystem_device == - PCI_SUBDEVICE_ID_QEMU)) { - dev_priv->pch_type = - intel_virt_detect_pch(dev_priv); - } else - continue; - - break; + unsigned short id; + + if (pch->vendor != PCI_VENDOR_ID_INTEL) + continue; + + id = pch->device & INTEL_PCH_DEVICE_ID_MASK; + + dev_priv->pch_id = id; + + if (id == INTEL_PCH_IBX_DEVICE_ID_TYPE) { + dev_priv->pch_type = PCH_IBX; + DRM_DEBUG_KMS("Found Ibex Peak PCH\n"); + WARN_ON(!IS_GEN5(dev_priv)); + } else if (id == INTEL_PCH_CPT_DEVICE_ID_TYPE) { + dev_priv->pch_type = PCH_CPT; + DRM_DEBUG_KMS("Found CougarPoint PCH\n"); + WARN_ON(!IS_GEN6(dev_priv) && + !IS_IVYBRIDGE(dev_priv)); + } else if (id == INTEL_PCH_PPT_DEVICE_ID_TYPE) { + /* PantherPoint is CPT compatible */ + dev_priv->pch_type = PCH_CPT; + DRM_DEBUG_KMS("Found PantherPoint PCH\n"); + WARN_ON(!IS_GEN6(dev_priv) && + !IS_IVYBRIDGE(dev_priv)); + } else if (id == INTEL_PCH_LPT_DEVICE_ID_TYPE) { + dev_priv->pch_type = PCH_LPT; + DRM_DEBUG_KMS("Found LynxPoint PCH\n"); + WARN_ON(!IS_HASWELL(dev_priv) && + !IS_BROADWELL(dev_priv)); + WARN_ON(IS_HSW_ULT(dev_priv) || + IS_BDW_ULT(dev_priv)); + } else if (id == INTEL_PCH_LPT_LP_DEVICE_ID_TYPE) { + dev_priv->pch_type = PCH_LPT; + DRM_DEBUG_KMS("Found LynxPoint LP PCH\n"); + WARN_ON(!IS_HASWELL(dev_priv) && + !IS_BROADWELL(dev_priv)); + WARN_ON(!IS_HSW_ULT(dev_priv) && + !IS_BDW_ULT(dev_priv)); + } else if (id == INTEL_PCH_WPT_DEVICE_ID_TYPE) { + /* WildcatPoint is LPT compatible */ + dev_priv->pch_type = PCH_LPT; + DRM_DEBUG_KMS("Found WildcatPoint PCH\n"); + WARN_ON(!IS_HASWELL(dev_priv) && + !IS_BROADWELL(dev_priv)); + WARN_ON(IS_HSW_ULT(dev_priv) || + IS_BDW_ULT(dev_priv)); + } else if (id == INTEL_PCH_WPT_LP_DEVICE_ID_TYPE) { + /* WildcatPoint is LPT compatible */ + dev_priv->pch_type = PCH_LPT; + DRM_DEBUG_KMS("Found WildcatPoint LP PCH\n"); + WARN_ON(!IS_HASWELL(dev_priv) && + !IS_BROADWELL(dev_priv)); + WARN_ON(!IS_HSW_ULT(dev_priv) && + !IS_BDW_ULT(dev_priv)); + } else if (id == INTEL_PCH_SPT_DEVICE_ID_TYPE) { + dev_priv->pch_type = PCH_SPT; + DRM_DEBUG_KMS("Found SunrisePoint PCH\n"); + WARN_ON(!IS_SKYLAKE(dev_priv) && + !IS_KABYLAKE(dev_priv)); + } else if (id == INTEL_PCH_SPT_LP_DEVICE_ID_TYPE) { + dev_priv->pch_type = PCH_SPT; + DRM_DEBUG_KMS("Found SunrisePoint LP PCH\n"); + WARN_ON(!IS_SKYLAKE(dev_priv) && + !IS_KABYLAKE(dev_priv)); + } else if (id == INTEL_PCH_KBP_DEVICE_ID_TYPE) { + dev_priv->pch_type = PCH_KBP; + DRM_DEBUG_KMS("Found Kaby Lake PCH (KBP)\n"); + WARN_ON(!IS_SKYLAKE(dev_priv) && + !IS_KABYLAKE(dev_priv) && + !IS_COFFEELAKE(dev_priv)); + } else if (id == INTEL_PCH_CNP_DEVICE_ID_TYPE) { + dev_priv->pch_type = PCH_CNP; + DRM_DEBUG_KMS("Found Cannon Lake PCH (CNP)\n"); + WARN_ON(!IS_CANNONLAKE(dev_priv) && + !IS_COFFEELAKE(dev_priv)); + } else if (id == INTEL_PCH_CNP_LP_DEVICE_ID_TYPE) { + dev_priv->pch_type = PCH_CNP; + DRM_DEBUG_KMS("Found Cannon Lake LP PCH (CNP-LP)\n"); + WARN_ON(!IS_CANNONLAKE(dev_priv) && + !IS_COFFEELAKE(dev_priv)); + } else if (id == INTEL_PCH_ICP_DEVICE_ID_TYPE) { + dev_priv->pch_type = PCH_ICP; + DRM_DEBUG_KMS("Found Ice Lake PCH\n"); + WARN_ON(!IS_ICELAKE(dev_priv)); + } else if (id == INTEL_PCH_P2X_DEVICE_ID_TYPE || + id == INTEL_PCH_P3X_DEVICE_ID_TYPE || + (id == INTEL_PCH_QEMU_DEVICE_ID_TYPE && + pch->subsystem_vendor == + PCI_SUBVENDOR_ID_REDHAT_QUMRANET && + pch->subsystem_device == + PCI_SUBDEVICE_ID_QEMU)) { + dev_priv->pch_type = intel_virt_detect_pch(dev_priv); + } else { + continue; } + + break; } if (!pch) DRM_DEBUG_KMS("No PCH found.\n"); @@ -622,7 +635,7 @@ static void i915_gem_fini(struct drm_i915_private *dev_priv) i915_gem_contexts_fini(dev_priv); mutex_unlock(&dev_priv->drm.struct_mutex); - intel_uc_fini_wq(dev_priv); + intel_uc_fini_misc(dev_priv); i915_gem_cleanup_userptr(dev_priv); i915_gem_drain_freed_objects(dev_priv); @@ -2596,6 +2609,11 @@ static int intel_runtime_suspend(struct device *kdev) intel_runtime_pm_enable_interrupts(dev_priv); + intel_guc_resume(dev_priv); + + i915_gem_init_swizzling(dev_priv); + i915_gem_restore_fences(dev_priv); + enable_rpm_wakeref_asserts(dev_priv); return ret; @@ -2661,8 +2679,6 @@ static int intel_runtime_resume(struct device *kdev) if (intel_uncore_unclaimed_mmio(dev_priv)) DRM_DEBUG_DRIVER("Unclaimed access during suspend, bios?\n"); - intel_guc_resume(dev_priv); - if (IS_GEN9_LP(dev_priv)) { bxt_disable_dc9(dev_priv); bxt_display_core_init(dev_priv, true); @@ -2677,6 +2693,10 @@ static int intel_runtime_resume(struct device *kdev) intel_uncore_runtime_resume(dev_priv); + intel_runtime_pm_enable_interrupts(dev_priv); + + intel_guc_resume(dev_priv); + /* * No point of rolling back things in case of an error, as the best * we can do is to hope that things will still work (and disable RPM). @@ -2684,8 +2704,6 @@ static int intel_runtime_resume(struct device *kdev) i915_gem_init_swizzling(dev_priv); i915_gem_restore_fences(dev_priv); - intel_runtime_pm_enable_interrupts(dev_priv); - /* * On VLV/CHV display interrupts are part of the display * power well, so hpd is reinitialized from there. For diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index a42deebedb0f..60079a69c2f9 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -83,8 +83,8 @@ #define DRIVER_NAME "i915" #define DRIVER_DESC "Intel Graphics" -#define DRIVER_DATE "20171222" -#define DRIVER_TIMESTAMP 1513971710 +#define DRIVER_DATE "20180207" +#define DRIVER_TIMESTAMP 1517988364 /* Use I915_STATE_WARN(x) and I915_STATE_WARN_ON() (rather than WARN() and * WARN_ON()) for hw state sanity checks to check for unexpected conditions @@ -104,9 +104,13 @@ #define I915_STATE_WARN_ON(x) \ I915_STATE_WARN((x), "%s", "WARN_ON(" __stringify(x) ")") +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) bool __i915_inject_load_failure(const char *func, int line); #define i915_inject_load_failure() \ __i915_inject_load_failure(__func__, __LINE__) +#else +#define i915_inject_load_failure() false +#endif typedef struct { uint32_t val; @@ -453,9 +457,9 @@ struct intel_display_error_state; struct i915_gpu_state { struct kref ref; - struct timeval time; - struct timeval boottime; - struct timeval uptime; + ktime_t time; + ktime_t boottime; + ktime_t uptime; struct drm_i915_private *i915; @@ -551,6 +555,7 @@ struct i915_gpu_state { int ban_score; int active; int guilty; + bool bannable; } context; struct drm_i915_error_object { @@ -754,7 +759,6 @@ struct i915_drrs { struct i915_psr { struct mutex lock; bool sink_support; - bool source_ok; struct intel_dp *enabled; bool active; struct delayed_work work; @@ -783,6 +787,7 @@ enum intel_pch { PCH_SPT, /* Sunrisepoint PCH */ PCH_KBP, /* Kaby Lake PCH */ PCH_CNP, /* Cannon Lake PCH */ + PCH_ICP, /* Ice Lake PCH */ PCH_NOP, }; @@ -1255,6 +1260,7 @@ enum modeset_restore { #define DP_AUX_B 0x10 #define DP_AUX_C 0x20 #define DP_AUX_D 0x30 +#define DP_AUX_F 0x60 #define DDC_PIN_B 0x05 #define DDC_PIN_C 0x04 @@ -1281,6 +1287,7 @@ struct ddi_vbt_port_info { uint8_t dp_boost_level; uint8_t hdmi_boost_level; + int dp_max_link_rate; /* 0 for not limited by VBT */ }; enum psr_lines_to_wait { @@ -1460,6 +1467,7 @@ struct skl_wm_params { uint_fixed_16_16_t plane_blocks_per_line; uint_fixed_16_16_t y_tile_minimum; uint32_t linetime_us; + uint32_t dbuf_block_size; }; /* @@ -1792,7 +1800,7 @@ struct i915_oa_ops { }; struct intel_cdclk_state { - unsigned int cdclk, vco, ref; + unsigned int cdclk, vco, ref, bypass; u8 voltage_level; }; @@ -2313,6 +2321,12 @@ struct drm_i915_private { bool awake; /** + * The number of times we have woken up. + */ + unsigned int epoch; +#define I915_EPOCH_INVALID 0 + + /** * We leave the user IRQ off as much as possible, * but this means that requests will finish and never * be retired once the system goes idle. Set a timer to @@ -2404,16 +2418,11 @@ enum hdmi_force_audio { * * We have one bit per pipe and per scanout plane type. */ -#define INTEL_MAX_SPRITE_BITS_PER_PIPE 5 #define INTEL_FRONTBUFFER_BITS_PER_PIPE 8 -#define INTEL_FRONTBUFFER_PRIMARY(pipe) \ - (1 << (INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe))) -#define INTEL_FRONTBUFFER_CURSOR(pipe) \ - (1 << (1 + (INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe)))) -#define INTEL_FRONTBUFFER_SPRITE(pipe, plane) \ - (1 << (2 + plane + (INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe)))) +#define INTEL_FRONTBUFFER(pipe, plane_id) \ + (1 << ((plane_id) + INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe))) #define INTEL_FRONTBUFFER_OVERLAY(pipe) \ - (1 << (2 + INTEL_MAX_SPRITE_BITS_PER_PIPE + (INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe)))) + (1 << (INTEL_FRONTBUFFER_BITS_PER_PIPE - 1 + INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe))) #define INTEL_FRONTBUFFER_ALL_MASK(pipe) \ (0xff << (INTEL_FRONTBUFFER_BITS_PER_PIPE * (pipe))) @@ -2595,6 +2604,7 @@ intel_info(const struct drm_i915_private *dev_priv) #define IS_GEMINILAKE(dev_priv) IS_PLATFORM(dev_priv, INTEL_GEMINILAKE) #define IS_COFFEELAKE(dev_priv) IS_PLATFORM(dev_priv, INTEL_COFFEELAKE) #define IS_CANNONLAKE(dev_priv) IS_PLATFORM(dev_priv, INTEL_CANNONLAKE) +#define IS_ICELAKE(dev_priv) IS_PLATFORM(dev_priv, INTEL_ICELAKE) #define IS_MOBILE(dev_priv) ((dev_priv)->info.is_mobile) #define IS_HSW_EARLY_SDV(dev_priv) (IS_HASWELL(dev_priv) && \ (INTEL_DEVID(dev_priv) & 0xFF00) == 0x0C00) @@ -2646,6 +2656,8 @@ intel_info(const struct drm_i915_private *dev_priv) (dev_priv)->info.gt == 2) #define IS_CFL_GT3(dev_priv) (IS_COFFEELAKE(dev_priv) && \ (dev_priv)->info.gt == 3) +#define IS_CNL_WITH_PORT_F(dev_priv) (IS_CANNONLAKE(dev_priv) && \ + (INTEL_DEVID(dev_priv) & 0x0004) == 0x0004) #define IS_ALPHA_SUPPORT(intel_info) ((intel_info)->is_alpha_support) @@ -2706,6 +2718,7 @@ intel_info(const struct drm_i915_private *dev_priv) #define IS_GEN8(dev_priv) (!!((dev_priv)->info.gen_mask & BIT(7))) #define IS_GEN9(dev_priv) (!!((dev_priv)->info.gen_mask & BIT(8))) #define IS_GEN10(dev_priv) (!!((dev_priv)->info.gen_mask & BIT(9))) +#define IS_GEN11(dev_priv) (!!((dev_priv)->info.gen_mask & BIT(10))) #define IS_LP(dev_priv) (INTEL_INFO(dev_priv)->is_lp) #define IS_GEN9_LP(dev_priv) (IS_GEN9(dev_priv) && IS_LP(dev_priv)) @@ -2843,11 +2856,13 @@ intel_info(const struct drm_i915_private *dev_priv) #define INTEL_PCH_KBP_DEVICE_ID_TYPE 0xA280 #define INTEL_PCH_CNP_DEVICE_ID_TYPE 0xA300 #define INTEL_PCH_CNP_LP_DEVICE_ID_TYPE 0x9D80 +#define INTEL_PCH_ICP_DEVICE_ID_TYPE 0x3480 #define INTEL_PCH_P2X_DEVICE_ID_TYPE 0x7100 #define INTEL_PCH_P3X_DEVICE_ID_TYPE 0x7000 #define INTEL_PCH_QEMU_DEVICE_ID_TYPE 0x2900 /* qemu q35 has 2918 */ #define INTEL_PCH_TYPE(dev_priv) ((dev_priv)->pch_type) +#define HAS_PCH_ICP(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_ICP) #define HAS_PCH_CNP(dev_priv) (INTEL_PCH_TYPE(dev_priv) == PCH_CNP) #define HAS_PCH_CNP_LP(dev_priv) \ ((dev_priv)->pch_id == INTEL_PCH_CNP_LP_DEVICE_ID_TYPE) @@ -2950,8 +2965,10 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv, void intel_hpd_init(struct drm_i915_private *dev_priv); void intel_hpd_init_work(struct drm_i915_private *dev_priv); void intel_hpd_cancel_work(struct drm_i915_private *dev_priv); -enum port intel_hpd_pin_to_port(enum hpd_pin pin); -enum hpd_pin intel_hpd_pin(enum port port); +enum port intel_hpd_pin_to_port(struct drm_i915_private *dev_priv, + enum hpd_pin pin); +enum hpd_pin intel_hpd_pin_default(struct drm_i915_private *dev_priv, + enum port port); bool intel_hpd_disable(struct drm_i915_private *dev_priv, enum hpd_pin pin); void intel_hpd_enable(struct drm_i915_private *dev_priv, enum hpd_pin pin); @@ -3718,9 +3735,10 @@ extern void intel_display_print_error_state(struct drm_i915_error_state_buf *e, int sandybridge_pcode_read(struct drm_i915_private *dev_priv, u32 mbox, u32 *val); int sandybridge_pcode_write_timeout(struct drm_i915_private *dev_priv, u32 mbox, - u32 val, int timeout_us); + u32 val, int fast_timeout_us, + int slow_timeout_ms); #define sandybridge_pcode_write(dev_priv, mbox, val) \ - sandybridge_pcode_write_timeout(dev_priv, mbox, val, 500) + sandybridge_pcode_write_timeout(dev_priv, mbox, val, 500, 0) int skl_pcode_request(struct drm_i915_private *dev_priv, u32 mbox, u32 request, u32 reply_mask, u32 reply, int timeout_base_ms); diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index dd89abd2263d..1632f18e6a64 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -369,7 +369,8 @@ i915_gem_object_wait_fence(struct dma_fence *fence, if (i915_gem_request_completed(rq)) goto out; - /* This client is about to stall waiting for the GPU. In many cases + /* + * This client is about to stall waiting for the GPU. In many cases * this is undesirable and limits the throughput of the system, as * many clients cannot continue processing user input/output whilst * blocked. RPS autotuning may take tens of milliseconds to respond @@ -384,11 +385,9 @@ i915_gem_object_wait_fence(struct dma_fence *fence, * forcing the clocks too high for the whole system, we only allow * each client to waitboost once in a busy period. */ - if (rps_client) { + if (rps_client && !i915_gem_request_started(rq)) { if (INTEL_GEN(rq->i915) >= 6) gen6_rps_boost(rq, rps_client); - else - rps_client = NULL; } timeout = i915_wait_request(rq, flags, timeout); @@ -2824,24 +2823,23 @@ i915_gem_object_pwrite_gtt(struct drm_i915_gem_object *obj, return 0; } -static bool ban_context(const struct i915_gem_context *ctx, - unsigned int score) -{ - return (i915_gem_context_is_bannable(ctx) && - score >= CONTEXT_SCORE_BAN_THRESHOLD); -} - static void i915_gem_context_mark_guilty(struct i915_gem_context *ctx) { - unsigned int score; bool banned; atomic_inc(&ctx->guilty_count); - score = atomic_add_return(CONTEXT_SCORE_GUILTY, &ctx->ban_score); - banned = ban_context(ctx, score); - DRM_DEBUG_DRIVER("context %s marked guilty (score %d) banned? %s\n", - ctx->name, score, yesno(banned)); + banned = false; + if (i915_gem_context_is_bannable(ctx)) { + unsigned int score; + + score = atomic_add_return(CONTEXT_SCORE_GUILTY, + &ctx->ban_score); + banned = score >= CONTEXT_SCORE_BAN_THRESHOLD; + + DRM_DEBUG_DRIVER("context %s marked guilty (score %d) banned? %s\n", + ctx->name, score, yesno(banned)); + } if (!banned) return; @@ -3135,7 +3133,7 @@ void i915_gem_reset(struct drm_i915_private *dev_priv) * an incoherent read by the CS (presumably stale TLB). An * empty request appears sufficient to paper over the glitch. */ - if (list_empty(&engine->timeline->requests)) { + if (intel_engine_is_idle(engine)) { struct drm_i915_gem_request *rq; rq = i915_gem_request_alloc(engine, @@ -3200,6 +3198,13 @@ void i915_gem_set_wedged(struct drm_i915_private *i915) struct intel_engine_cs *engine; enum intel_engine_id id; + if (drm_debug & DRM_UT_DRIVER) { + struct drm_printer p = drm_debug_printer(__func__); + + for_each_engine(engine, i915, id) + intel_engine_dump(engine, &p, "%s\n", engine->name); + } + /* * First, stop submission to hw, but do not yet complete requests by * rolling the global seqno forward (since this would complete requests @@ -3334,6 +3339,65 @@ i915_gem_retire_work_handler(struct work_struct *work) round_jiffies_up_relative(HZ)); } +static void shrink_caches(struct drm_i915_private *i915) +{ + /* + * kmem_cache_shrink() discards empty slabs and reorders partially + * filled slabs to prioritise allocating from the mostly full slabs, + * with the aim of reducing fragmentation. + */ + kmem_cache_shrink(i915->priorities); + kmem_cache_shrink(i915->dependencies); + kmem_cache_shrink(i915->requests); + kmem_cache_shrink(i915->luts); + kmem_cache_shrink(i915->vmas); + kmem_cache_shrink(i915->objects); +} + +struct sleep_rcu_work { + union { + struct rcu_head rcu; + struct work_struct work; + }; + struct drm_i915_private *i915; + unsigned int epoch; +}; + +static inline bool +same_epoch(struct drm_i915_private *i915, unsigned int epoch) +{ + /* + * There is a small chance that the epoch wrapped since we started + * sleeping. If we assume that epoch is at least a u32, then it will + * take at least 2^32 * 100ms for it to wrap, or about 326 years. + */ + return epoch == READ_ONCE(i915->gt.epoch); +} + +static void __sleep_work(struct work_struct *work) +{ + struct sleep_rcu_work *s = container_of(work, typeof(*s), work); + struct drm_i915_private *i915 = s->i915; + unsigned int epoch = s->epoch; + + kfree(s); + if (same_epoch(i915, epoch)) + shrink_caches(i915); +} + +static void __sleep_rcu(struct rcu_head *rcu) +{ + struct sleep_rcu_work *s = container_of(rcu, typeof(*s), rcu); + struct drm_i915_private *i915 = s->i915; + + if (same_epoch(i915, s->epoch)) { + INIT_WORK(&s->work, __sleep_work); + queue_work(i915->wq, &s->work); + } else { + kfree(s); + } +} + static inline bool new_requests_since_last_retire(const struct drm_i915_private *i915) { @@ -3346,6 +3410,7 @@ i915_gem_idle_work_handler(struct work_struct *work) { struct drm_i915_private *dev_priv = container_of(work, typeof(*dev_priv), gt.idle_work.work); + unsigned int epoch = I915_EPOCH_INVALID; bool rearm_hangcheck; ktime_t end; @@ -3405,6 +3470,8 @@ i915_gem_idle_work_handler(struct work_struct *work) GEM_BUG_ON(!dev_priv->gt.awake); dev_priv->gt.awake = false; + epoch = dev_priv->gt.epoch; + GEM_BUG_ON(epoch == I915_EPOCH_INVALID); rearm_hangcheck = false; if (INTEL_GEN(dev_priv) >= 6) @@ -3421,6 +3488,23 @@ out_rearm: GEM_BUG_ON(!dev_priv->gt.awake); i915_queue_hangcheck(dev_priv); } + + /* + * When we are idle, it is an opportune time to reap our caches. + * However, we have many objects that utilise RCU and the ordered + * i915->wq that this work is executing on. To try and flush any + * pending frees now we are idle, we first wait for an RCU grace + * period, and then queue a task (that will run last on the wq) to + * shrink and re-optimize the caches. + */ + if (same_epoch(dev_priv, epoch)) { + struct sleep_rcu_work *s = kmalloc(sizeof(*s), GFP_KERNEL); + if (s) { + s->i915 = dev_priv; + s->epoch = epoch; + call_rcu(&s->rcu, __sleep_rcu); + } + } } void i915_gem_close_object(struct drm_gem_object *gem, struct drm_file *file) @@ -3566,7 +3650,7 @@ static int wait_for_engines(struct drm_i915_private *i915) for_each_engine(engine, i915, id) intel_engine_dump(engine, &p, - "%s", engine->name); + "%s\n", engine->name); } i915_gem_set_wedged(i915); @@ -4698,7 +4782,8 @@ static void __i915_gem_free_work(struct work_struct *work) container_of(work, struct drm_i915_private, mm.free_work); struct llist_node *freed; - /* All file-owned VMA should have been released by this point through + /* + * All file-owned VMA should have been released by this point through * i915_gem_close_object(), or earlier by i915_gem_context_close(). * However, the object may also be bound into the global GTT (e.g. * older GPUs without per-process support, or for direct access through @@ -4725,13 +4810,18 @@ static void __i915_gem_free_object_rcu(struct rcu_head *head) container_of(head, typeof(*obj), rcu); struct drm_i915_private *i915 = to_i915(obj->base.dev); - /* We can't simply use call_rcu() from i915_gem_free_object() - * as we need to block whilst unbinding, and the call_rcu - * task may be called from softirq context. So we take a - * detour through a worker. + /* + * Since we require blocking on struct_mutex to unbind the freed + * object from the GPU before releasing resources back to the + * system, we can not do that directly from the RCU callback (which may + * be a softirq context), but must instead then defer that work onto a + * kthread. We use the RCU callback rather than move the freed object + * directly onto the work queue so that we can mix between using the + * worker and performing frees directly from subsequent allocations for + * crude but effective memory throttling. */ if (llist_add(&obj->freed, &i915->mm.free_list)) - schedule_work(&i915->mm.free_work); + queue_work(i915->wq, &i915->mm.free_work); } void i915_gem_free_object(struct drm_gem_object *gem_obj) @@ -4744,7 +4834,8 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj) if (discard_backing_storage(obj)) obj->mm.madv = I915_MADV_DONTNEED; - /* Before we free the object, make sure any pure RCU-only + /* + * Before we free the object, make sure any pure RCU-only * read-side critical sections are complete, e.g. * i915_gem_busy_ioctl(). For the corresponding synchronized * lookup see i915_gem_object_lookup_rcu(). @@ -5186,7 +5277,7 @@ int i915_gem_init(struct drm_i915_private *dev_priv) if (ret) return ret; - ret = intel_uc_init_wq(dev_priv); + ret = intel_uc_init_misc(dev_priv); if (ret) return ret; @@ -5282,7 +5373,7 @@ err_unlock: intel_uncore_forcewake_put(dev_priv, FORCEWAKE_ALL); mutex_unlock(&dev_priv->drm.struct_mutex); - intel_uc_fini_wq(dev_priv); + intel_uc_fini_misc(dev_priv); if (ret != -EIO) i915_gem_cleanup_userptr(dev_priv); diff --git a/drivers/gpu/drm/i915/i915_gem_fence_reg.c b/drivers/gpu/drm/i915/i915_gem_fence_reg.c index 012250f25255..b8338d75c6f3 100644 --- a/drivers/gpu/drm/i915/i915_gem_fence_reg.c +++ b/drivers/gpu/drm/i915/i915_gem_fence_reg.c @@ -230,10 +230,14 @@ static int fence_update(struct drm_i915_fence_reg *fence, } if (fence->vma) { - ret = i915_gem_active_retire(&fence->vma->last_fence, - &fence->vma->obj->base.dev->struct_mutex); + struct i915_vma *old = fence->vma; + + ret = i915_gem_active_retire(&old->last_fence, + &old->obj->base.dev->struct_mutex); if (ret) return ret; + + i915_vma_flush_writes(old); } if (fence->vma && fence->vma != vma) { diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 7e403eaa9e0f..955ce7bee448 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -543,9 +543,7 @@ static void fill_page_dma_32(struct i915_address_space *vm, static int setup_scratch_page(struct i915_address_space *vm, gfp_t gfp) { - struct page *page = NULL; - dma_addr_t addr; - int order; + unsigned long size; /* * In order to utilize 64K pages for an object with a size < 2M, we will @@ -559,48 +557,47 @@ setup_scratch_page(struct i915_address_space *vm, gfp_t gfp) * TODO: we should really consider write-protecting the scratch-page and * sharing between ppgtt */ + size = I915_GTT_PAGE_SIZE_4K; if (i915_vm_is_48bit(vm) && HAS_PAGE_SIZES(vm->i915, I915_GTT_PAGE_SIZE_64K)) { - order = get_order(I915_GTT_PAGE_SIZE_64K); - page = alloc_pages(gfp | __GFP_ZERO | __GFP_NOWARN, order); - if (page) { - addr = dma_map_page(vm->dma, page, 0, - I915_GTT_PAGE_SIZE_64K, - PCI_DMA_BIDIRECTIONAL); - if (unlikely(dma_mapping_error(vm->dma, addr))) { - __free_pages(page, order); - page = NULL; - } - - if (!IS_ALIGNED(addr, I915_GTT_PAGE_SIZE_64K)) { - dma_unmap_page(vm->dma, addr, - I915_GTT_PAGE_SIZE_64K, - PCI_DMA_BIDIRECTIONAL); - __free_pages(page, order); - page = NULL; - } - } + size = I915_GTT_PAGE_SIZE_64K; + gfp |= __GFP_NOWARN; } + gfp |= __GFP_ZERO | __GFP_RETRY_MAYFAIL; - if (!page) { - order = 0; - page = alloc_page(gfp | __GFP_ZERO); + do { + int order = get_order(size); + struct page *page; + dma_addr_t addr; + + page = alloc_pages(gfp, order); if (unlikely(!page)) - return -ENOMEM; + goto skip; - addr = dma_map_page(vm->dma, page, 0, PAGE_SIZE, + addr = dma_map_page(vm->dma, page, 0, size, PCI_DMA_BIDIRECTIONAL); - if (unlikely(dma_mapping_error(vm->dma, addr))) { - __free_page(page); - return -ENOMEM; - } - } + if (unlikely(dma_mapping_error(vm->dma, addr))) + goto free_page; - vm->scratch_page.page = page; - vm->scratch_page.daddr = addr; - vm->scratch_page.order = order; + if (unlikely(!IS_ALIGNED(addr, size))) + goto unmap_page; - return 0; + vm->scratch_page.page = page; + vm->scratch_page.daddr = addr; + vm->scratch_page.order = order; + return 0; + +unmap_page: + dma_unmap_page(vm->dma, addr, size, PCI_DMA_BIDIRECTIONAL); +free_page: + __free_pages(page, order); +skip: + if (size == I915_GTT_PAGE_SIZE_4K) + return -ENOMEM; + + size = I915_GTT_PAGE_SIZE_4K; + gfp &= ~__GFP_NOWARN; + } while (1); } static void cleanup_scratch_page(struct i915_address_space *vm) @@ -2370,9 +2367,10 @@ int i915_gem_gtt_prepare_pages(struct drm_i915_gem_object *obj, struct sg_table *pages) { do { - if (dma_map_sg(&obj->base.dev->pdev->dev, - pages->sgl, pages->nents, - PCI_DMA_BIDIRECTIONAL)) + if (dma_map_sg_attrs(&obj->base.dev->pdev->dev, + pages->sgl, pages->nents, + PCI_DMA_BIDIRECTIONAL, + DMA_ATTR_NO_WARN)) return 0; /* If the DMA remap fails, one cause can be that we have diff --git a/drivers/gpu/drm/i915/i915_gem_request.c b/drivers/gpu/drm/i915/i915_gem_request.c index e09d18df8b7f..8efa9e7a9e46 100644 --- a/drivers/gpu/drm/i915/i915_gem_request.c +++ b/drivers/gpu/drm/i915/i915_gem_request.c @@ -161,12 +161,16 @@ i915_priotree_fini(struct drm_i915_private *i915, struct i915_priotree *pt) GEM_BUG_ON(!list_empty(&pt->link)); - /* Everyone we depended upon (the fences we wait to be signaled) + /* + * Everyone we depended upon (the fences we wait to be signaled) * should retire before us and remove themselves from our list. * However, retirement is run independently on each timeline and * so we may be called out-of-order. */ list_for_each_entry_safe(dep, next, &pt->signalers_list, signal_link) { + GEM_BUG_ON(!i915_priotree_signaled(dep->signaler)); + GEM_BUG_ON(!list_empty(&dep->dfs_link)); + list_del(&dep->wait_link); if (dep->flags & I915_DEPENDENCY_ALLOC) i915_dependency_free(i915, dep); @@ -174,6 +178,9 @@ i915_priotree_fini(struct drm_i915_private *i915, struct i915_priotree *pt) /* Remove ourselves from everyone who depends upon us */ list_for_each_entry_safe(dep, next, &pt->waiters_list, wait_link) { + GEM_BUG_ON(dep->signaler != pt); + GEM_BUG_ON(!list_empty(&dep->dfs_link)); + list_del(&dep->signal_link); if (dep->flags & I915_DEPENDENCY_ALLOC) i915_dependency_free(i915, dep); @@ -267,6 +274,8 @@ static void mark_busy(struct drm_i915_private *i915) intel_display_power_get(i915, POWER_DOMAIN_GT_IRQ); i915->gt.awake = true; + if (unlikely(++i915->gt.epoch == 0)) /* keep 0 as invalid */ + i915->gt.epoch = 1; intel_enable_gt_powersave(i915); i915_update_gfx_val(i915); @@ -436,7 +445,10 @@ static void i915_gem_request_retire(struct drm_i915_gem_request *request) spin_lock_irq(&request->lock); if (request->waitboost) atomic_dec(&request->i915->gt_pm.rps.num_waiters); - dma_fence_signal_locked(&request->fence); + if (!test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, &request->fence.flags)) + dma_fence_signal_locked(&request->fence); + if (test_bit(DMA_FENCE_FLAG_ENABLE_SIGNAL_BIT, &request->fence.flags)) + intel_engine_cancel_signaling(request); spin_unlock_irq(&request->lock); i915_priotree_fini(request->i915, &request->priotree); @@ -530,6 +542,8 @@ void __i915_gem_request_unsubmit(struct drm_i915_gem_request *request) */ GEM_BUG_ON(!request->global_seqno); GEM_BUG_ON(request->global_seqno != engine->timeline->seqno); + GEM_BUG_ON(i915_seqno_passed(intel_engine_get_seqno(engine), + request->global_seqno)); engine->timeline->seqno--; /* We may be recursing from the signal callback of another i915 fence */ @@ -691,6 +705,17 @@ i915_gem_request_alloc(struct intel_engine_cs *engine, if (ret) goto err_unreserve; + /* + * We've forced the client to stall and catch up with whatever + * backlog there might have been. As we are assuming that we + * caused the mempressure, now is an opportune time to + * recover as much memory from the request pool as is possible. + * Having already penalized the client to stall, we spend + * a little extra time to re-optimise page allocation. + */ + kmem_cache_shrink(dev_priv->requests); + rcu_barrier(); /* Recover the TYPESAFE_BY_RCU pages */ + req = kmem_cache_alloc(dev_priv->requests, GFP_KERNEL); if (!req) { ret = -ENOMEM; @@ -722,6 +747,7 @@ i915_gem_request_alloc(struct intel_engine_cs *engine, /* No zalloc, must clear what we need by hand */ req->global_seqno = 0; + req->signaling.wait.seqno = 0; req->file_priv = NULL; req->batch = NULL; req->capture_list = NULL; diff --git a/drivers/gpu/drm/i915/i915_gem_request.h b/drivers/gpu/drm/i915/i915_gem_request.h index 0d6d39f19506..2236e9188c5c 100644 --- a/drivers/gpu/drm/i915/i915_gem_request.h +++ b/drivers/gpu/drm/i915/i915_gem_request.h @@ -245,18 +245,6 @@ i915_gem_request_put(struct drm_i915_gem_request *req) dma_fence_put(&req->fence); } -static inline void i915_gem_request_assign(struct drm_i915_gem_request **pdst, - struct drm_i915_gem_request *src) -{ - if (src) - i915_gem_request_get(src); - - if (*pdst) - i915_gem_request_put(*pdst); - - *pdst = src; -} - /** * i915_gem_request_global_seqno - report the current global seqno * @request - the request @@ -341,6 +329,27 @@ i915_gem_request_completed(const struct drm_i915_gem_request *req) return __i915_gem_request_completed(req, seqno); } +static inline bool +i915_gem_request_started(const struct drm_i915_gem_request *req) +{ + u32 seqno; + + seqno = i915_gem_request_global_seqno(req); + if (!seqno) + return false; + + return i915_seqno_passed(intel_engine_get_seqno(req->engine), + seqno - 1); +} + +static inline bool i915_priotree_signaled(const struct i915_priotree *pt) +{ + const struct drm_i915_gem_request *rq = + container_of(pt, const struct drm_i915_gem_request, priotree); + + return i915_gem_request_completed(rq); +} + /* We treat requests as fences. This is not be to confused with our * "fence registers" but pipeline synchronisation objects ala GL_ARB_sync. * We use the fences to synchronize access from the CPU with activity on the diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index 944059322daa..67c902412193 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -34,16 +34,25 @@ #include "i915_drv.h" -static const char *engine_str(int engine) -{ - switch (engine) { - case RCS: return "render"; - case VCS: return "bsd"; - case BCS: return "blt"; - case VECS: return "vebox"; - case VCS2: return "bsd2"; - default: return ""; - } +static inline const struct intel_engine_cs * +engine_lookup(const struct drm_i915_private *i915, unsigned int id) +{ + if (id >= I915_NUM_ENGINES) + return NULL; + + return i915->engine[id]; +} + +static inline const char * +__engine_name(const struct intel_engine_cs *engine) +{ + return engine ? engine->name : ""; +} + +static const char * +engine_name(const struct drm_i915_private *i915, unsigned int id) +{ + return __engine_name(engine_lookup(i915, id)); } static const char *tiling_flag(int tiling) @@ -345,7 +354,7 @@ static void print_error_buffers(struct drm_i915_error_state_buf *m, err_puts(m, purgeable_flag(err->purgeable)); err_puts(m, err->userptr ? " userptr" : ""); err_puts(m, err->engine != -1 ? " " : ""); - err_puts(m, engine_str(err->engine)); + err_puts(m, engine_name(m->i915, err->engine)); err_puts(m, i915_cache_level_str(m->i915, err->cache_level)); if (err->name) @@ -387,6 +396,11 @@ static void error_print_instdone(struct drm_i915_error_state_buf *m, ee->instdone.row[slice][subslice]); } +static const char *bannable(const struct drm_i915_error_context *ctx) +{ + return ctx->bannable ? "" : " (unbannable)"; +} + static void error_print_request(struct drm_i915_error_state_buf *m, const char *prefix, const struct drm_i915_error_request *erq) @@ -405,9 +419,10 @@ static void error_print_context(struct drm_i915_error_state_buf *m, const char *header, const struct drm_i915_error_context *ctx) { - err_printf(m, "%s%s[%d] user_handle %d hw_id %d, prio %d, ban score %d guilty %d active %d\n", + err_printf(m, "%s%s[%d] user_handle %d hw_id %d, prio %d, ban score %d%s guilty %d active %d\n", header, ctx->comm, ctx->pid, ctx->handle, ctx->hw_id, - ctx->priority, ctx->ban_score, ctx->guilty, ctx->active); + ctx->priority, ctx->ban_score, bannable(ctx), + ctx->guilty, ctx->active); } static void error_print_engine(struct drm_i915_error_state_buf *m, @@ -415,7 +430,8 @@ static void error_print_engine(struct drm_i915_error_state_buf *m, { int n; - err_printf(m, "%s command stream:\n", engine_str(ee->engine_id)); + err_printf(m, "%s command stream:\n", + engine_name(m->i915, ee->engine_id)); err_printf(m, " IDLE?: %s\n", yesno(ee->idle)); err_printf(m, " START: 0x%08x\n", ee->start); err_printf(m, " HEAD: 0x%08x [0x%08x]\n", ee->head, ee->rq_head); @@ -610,6 +626,7 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, { struct drm_i915_private *dev_priv = m->i915; struct drm_i915_error_object *obj; + struct timespec64 ts; int i, j; if (!error) { @@ -620,21 +637,25 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, if (*error->error_msg) err_printf(m, "%s\n", error->error_msg); err_printf(m, "Kernel: " UTS_RELEASE "\n"); - err_printf(m, "Time: %ld s %ld us\n", - error->time.tv_sec, error->time.tv_usec); - err_printf(m, "Boottime: %ld s %ld us\n", - error->boottime.tv_sec, error->boottime.tv_usec); - err_printf(m, "Uptime: %ld s %ld us\n", - error->uptime.tv_sec, error->uptime.tv_usec); + ts = ktime_to_timespec64(error->time); + err_printf(m, "Time: %lld s %ld us\n", + (s64)ts.tv_sec, ts.tv_nsec / NSEC_PER_USEC); + ts = ktime_to_timespec64(error->boottime); + err_printf(m, "Boottime: %lld s %ld us\n", + (s64)ts.tv_sec, ts.tv_nsec / NSEC_PER_USEC); + ts = ktime_to_timespec64(error->uptime); + err_printf(m, "Uptime: %lld s %ld us\n", + (s64)ts.tv_sec, ts.tv_nsec / NSEC_PER_USEC); for (i = 0; i < ARRAY_SIZE(error->engine); i++) { if (error->engine[i].hangcheck_stalled && error->engine[i].context.pid) { - err_printf(m, "Active process (on ring %s): %s [%d], score %d\n", - engine_str(i), + err_printf(m, "Active process (on ring %s): %s [%d], score %d%s\n", + engine_name(m->i915, i), error->engine[i].context.comm, error->engine[i].context.pid, - error->engine[i].context.ban_score); + error->engine[i].context.ban_score, + bannable(&error->engine[i].context)); } } err_printf(m, "Reset count: %u\n", error->reset_count); @@ -722,12 +743,13 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m, if (obj) { err_puts(m, dev_priv->engine[i]->name); if (ee->context.pid) - err_printf(m, " (submitted by %s [%d], ctx %d [%d], score %d)", + err_printf(m, " (submitted by %s [%d], ctx %d [%d], score %d%s)", ee->context.comm, ee->context.pid, ee->context.handle, ee->context.hw_id, - ee->context.ban_score); + ee->context.ban_score, + bannable(&ee->context)); err_printf(m, " --- gtt_offset = 0x%08x %08x\n", upper_32_bits(obj->gtt_offset), lower_32_bits(obj->gtt_offset)); @@ -1369,6 +1391,7 @@ static void record_context(struct drm_i915_error_context *e, e->hw_id = ctx->hw_id; e->priority = ctx->priority; e->ban_score = atomic_read(&ctx->ban_score); + e->bannable = i915_gem_context_is_bannable(ctx); e->guilty = atomic_read(&ctx->guilty_count); e->active = atomic_read(&ctx->active_count); } @@ -1737,11 +1760,10 @@ static int capture(void *data) { struct i915_gpu_state *error = data; - do_gettimeofday(&error->time); - error->boottime = ktime_to_timeval(ktime_get_boottime()); - error->uptime = - ktime_to_timeval(ktime_sub(ktime_get(), - error->i915->gt.last_init_time)); + error->time = ktime_get_real(); + error->boottime = ktime_get_boottime(); + error->uptime = ktime_sub(ktime_get(), + error->i915->gt.last_init_time); capture_params(error); capture_uc_state(error); diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 3517c6548e2c..b886bd459acc 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -452,6 +452,8 @@ void gen6_disable_rps_interrupts(struct drm_i915_private *dev_priv) void gen9_reset_guc_interrupts(struct drm_i915_private *dev_priv) { + assert_rpm_wakelock_held(dev_priv); + spin_lock_irq(&dev_priv->irq_lock); gen6_reset_pm_iir(dev_priv, dev_priv->pm_guc_events); spin_unlock_irq(&dev_priv->irq_lock); @@ -459,6 +461,8 @@ void gen9_reset_guc_interrupts(struct drm_i915_private *dev_priv) void gen9_enable_guc_interrupts(struct drm_i915_private *dev_priv) { + assert_rpm_wakelock_held(dev_priv); + spin_lock_irq(&dev_priv->irq_lock); if (!dev_priv->guc.interrupts_enabled) { WARN_ON_ONCE(I915_READ(gen6_pm_iir(dev_priv)) & @@ -471,6 +475,8 @@ void gen9_enable_guc_interrupts(struct drm_i915_private *dev_priv) void gen9_disable_guc_interrupts(struct drm_i915_private *dev_priv) { + assert_rpm_wakelock_held(dev_priv); + spin_lock_irq(&dev_priv->irq_lock); dev_priv->guc.interrupts_enabled = false; @@ -1407,37 +1413,25 @@ gen8_cs_irq_handler(struct intel_engine_cs *engine, u32 iir, int test_shift) tasklet_hi_schedule(&execlists->tasklet); } -static irqreturn_t gen8_gt_irq_ack(struct drm_i915_private *dev_priv, - u32 master_ctl, - u32 gt_iir[4]) +static void gen8_gt_irq_ack(struct drm_i915_private *dev_priv, + u32 master_ctl, u32 gt_iir[4]) { - irqreturn_t ret = IRQ_NONE; - if (master_ctl & (GEN8_GT_RCS_IRQ | GEN8_GT_BCS_IRQ)) { gt_iir[0] = I915_READ_FW(GEN8_GT_IIR(0)); - if (gt_iir[0]) { + if (gt_iir[0]) I915_WRITE_FW(GEN8_GT_IIR(0), gt_iir[0]); - ret = IRQ_HANDLED; - } else - DRM_ERROR("The master control interrupt lied (GT0)!\n"); } if (master_ctl & (GEN8_GT_VCS1_IRQ | GEN8_GT_VCS2_IRQ)) { gt_iir[1] = I915_READ_FW(GEN8_GT_IIR(1)); - if (gt_iir[1]) { + if (gt_iir[1]) I915_WRITE_FW(GEN8_GT_IIR(1), gt_iir[1]); - ret = IRQ_HANDLED; - } else - DRM_ERROR("The master control interrupt lied (GT1)!\n"); } if (master_ctl & GEN8_GT_VECS_IRQ) { gt_iir[3] = I915_READ_FW(GEN8_GT_IIR(3)); - if (gt_iir[3]) { + if (gt_iir[3]) I915_WRITE_FW(GEN8_GT_IIR(3), gt_iir[3]); - ret = IRQ_HANDLED; - } else - DRM_ERROR("The master control interrupt lied (GT3)!\n"); } if (master_ctl & (GEN8_GT_PM_IRQ | GEN8_GT_GUC_IRQ)) { @@ -1447,12 +1441,8 @@ static irqreturn_t gen8_gt_irq_ack(struct drm_i915_private *dev_priv, I915_WRITE_FW(GEN8_GT_IIR(2), gt_iir[2] & (dev_priv->pm_rps_events | dev_priv->pm_guc_events)); - ret = IRQ_HANDLED; - } else - DRM_ERROR("The master control interrupt lied (PM)!\n"); + } } - - return ret; } static void gen8_gt_irq_handler(struct drm_i915_private *dev_priv, @@ -1568,10 +1558,11 @@ static bool i9xx_port_hotplug_long_detect(enum port port, u32 val) * * Note that the caller is expected to zero out the masks initially. */ -static void intel_get_hpd_pins(u32 *pin_mask, u32 *long_mask, - u32 hotplug_trigger, u32 dig_hotplug_reg, - const u32 hpd[HPD_NUM_PINS], - bool long_pulse_detect(enum port port, u32 val)) +static void intel_get_hpd_pins(struct drm_i915_private *dev_priv, + u32 *pin_mask, u32 *long_mask, + u32 hotplug_trigger, u32 dig_hotplug_reg, + const u32 hpd[HPD_NUM_PINS], + bool long_pulse_detect(enum port port, u32 val)) { enum port port; int i; @@ -1582,7 +1573,7 @@ static void intel_get_hpd_pins(u32 *pin_mask, u32 *long_mask, *pin_mask |= BIT(i); - port = intel_hpd_pin_to_port(i); + port = intel_hpd_pin_to_port(dev_priv, i); if (port == PORT_NONE) continue; @@ -1970,8 +1961,9 @@ static void i9xx_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_G4X; if (hotplug_trigger) { - intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, - hotplug_trigger, hpd_status_g4x, + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + hotplug_trigger, hotplug_trigger, + hpd_status_g4x, i9xx_port_hotplug_long_detect); intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); @@ -1983,8 +1975,9 @@ static void i9xx_hpd_irq_handler(struct drm_i915_private *dev_priv, u32 hotplug_trigger = hotplug_status & HOTPLUG_INT_STATUS_I915; if (hotplug_trigger) { - intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, - hotplug_trigger, hpd_status_i915, + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + hotplug_trigger, hotplug_trigger, + hpd_status_i915, i9xx_port_hotplug_long_detect); intel_hpd_irq_handler(dev_priv, pin_mask, long_mask); } @@ -2185,7 +2178,7 @@ static void ibx_hpd_irq_handler(struct drm_i915_private *dev_priv, if (!hotplug_trigger) return; - intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, hotplug_trigger, dig_hotplug_reg, hpd, pch_port_hotplug_long_detect); @@ -2327,8 +2320,8 @@ static void spt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir) dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG); I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg); - intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, - dig_hotplug_reg, hpd_spt, + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + hotplug_trigger, dig_hotplug_reg, hpd_spt, spt_port_hotplug_long_detect); } @@ -2338,8 +2331,8 @@ static void spt_irq_handler(struct drm_i915_private *dev_priv, u32 pch_iir) dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG2); I915_WRITE(PCH_PORT_HOTPLUG2, dig_hotplug_reg); - intel_get_hpd_pins(&pin_mask, &long_mask, hotplug2_trigger, - dig_hotplug_reg, hpd_spt, + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, + hotplug2_trigger, dig_hotplug_reg, hpd_spt, spt_port_hotplug2_long_detect); } @@ -2359,7 +2352,7 @@ static void ilk_hpd_irq_handler(struct drm_i915_private *dev_priv, dig_hotplug_reg = I915_READ(DIGITAL_PORT_HOTPLUG_CNTRL); I915_WRITE(DIGITAL_PORT_HOTPLUG_CNTRL, dig_hotplug_reg); - intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, hotplug_trigger, dig_hotplug_reg, hpd, ilk_port_hotplug_long_detect); @@ -2536,7 +2529,7 @@ static void bxt_hpd_irq_handler(struct drm_i915_private *dev_priv, dig_hotplug_reg = I915_READ(PCH_PORT_HOTPLUG); I915_WRITE(PCH_PORT_HOTPLUG, dig_hotplug_reg); - intel_get_hpd_pins(&pin_mask, &long_mask, hotplug_trigger, + intel_get_hpd_pins(dev_priv, &pin_mask, &long_mask, hotplug_trigger, dig_hotplug_reg, hpd, bxt_port_hotplug_long_detect); @@ -2579,6 +2572,9 @@ gen8_de_irq_handler(struct drm_i915_private *dev_priv, u32 master_ctl) GEN9_AUX_CHANNEL_C | GEN9_AUX_CHANNEL_D; + if (IS_CNL_WITH_PORT_F(dev_priv)) + tmp_mask |= CNL_AUX_CHANNEL_F; + if (iir & tmp_mask) { dp_aux_irq_handler(dev_priv); found = true; @@ -2683,7 +2679,6 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg) struct drm_i915_private *dev_priv = to_i915(dev); u32 master_ctl; u32 gt_iir[4] = {}; - irqreturn_t ret; if (!intel_irqs_enabled(dev_priv)) return IRQ_NONE; @@ -2699,16 +2694,16 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg) disable_rpm_wakeref_asserts(dev_priv); /* Find, clear, then process each source of interrupt */ - ret = gen8_gt_irq_ack(dev_priv, master_ctl, gt_iir); + gen8_gt_irq_ack(dev_priv, master_ctl, gt_iir); gen8_gt_irq_handler(dev_priv, gt_iir); - ret |= gen8_de_irq_handler(dev_priv, master_ctl); + gen8_de_irq_handler(dev_priv, master_ctl); I915_WRITE_FW(GEN8_MASTER_IRQ, GEN8_MASTER_IRQ_CONTROL); POSTING_READ_FW(GEN8_MASTER_IRQ); enable_rpm_wakeref_asserts(dev_priv); - return ret; + return IRQ_HANDLED; } struct wedge_me { @@ -3611,6 +3606,9 @@ static void gen8_de_irq_postinstall(struct drm_i915_private *dev_priv) de_pipe_masked |= GEN8_DE_PIPE_IRQ_FAULT_ERRORS; } + if (IS_CNL_WITH_PORT_F(dev_priv)) + de_port_masked |= CNL_AUX_CHANNEL_F; + de_pipe_enables = de_pipe_masked | GEN8_PIPE_VBLANK | GEN8_PIPE_FIFO_UNDERRUN; diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c index b5f3eb4fa8a3..08108ce5be21 100644 --- a/drivers/gpu/drm/i915/i915_params.c +++ b/drivers/gpu/drm/i915/i915_params.c @@ -155,7 +155,8 @@ i915_param_named_unsafe(enable_guc, int, 0400, "(-1=auto, 0=disable [default], 1=GuC submission, 2=HuC load)"); i915_param_named(guc_log_level, int, 0400, - "GuC firmware logging level (-1:disabled (default), 0-3:enabled)"); + "GuC firmware logging level. Requires GuC to be loaded. " + "(-1=auto [default], 0=disable, 1..4=enable with verbosity min..max)"); i915_param_named_unsafe(guc_firmware_path, charp, 0400, "GuC firmware path to use instead of the default one"); @@ -166,8 +167,10 @@ i915_param_named_unsafe(huc_firmware_path, charp, 0400, i915_param_named_unsafe(enable_dp_mst, bool, 0600, "Enable multi-stream transport (MST) for new DisplayPort sinks. (default: true)"); +#if IS_ENABLED(CONFIG_DRM_I915_DEBUG) i915_param_named_unsafe(inject_load_failure, uint, 0400, "Force an error after a number of failure check points (0:disabled (default), N:force failure at the Nth failure check point)"); +#endif i915_param_named(enable_dpcd_backlight, bool, 0600, "Enable support for DPCD backlight control (default:false)"); diff --git a/drivers/gpu/drm/i915/i915_params.h b/drivers/gpu/drm/i915/i915_params.h index c96360398072..430f5f9d0ff4 100644 --- a/drivers/gpu/drm/i915/i915_params.h +++ b/drivers/gpu/drm/i915/i915_params.h @@ -48,7 +48,7 @@ struct drm_printer; param(int, enable_ips, 1) \ param(int, invert_brightness, 0) \ param(int, enable_guc, 0) \ - param(int, guc_log_level, -1) \ + param(int, guc_log_level, 0) \ param(char *, guc_firmware_path, NULL) \ param(char *, huc_firmware_path, NULL) \ param(int, mmio_debug, 0) \ diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index 1c30c688f23a..4e7a10c89782 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -571,7 +571,7 @@ static const struct intel_device_info intel_coffeelake_gt3_info = { .ddb_size = 1024, \ GLK_COLORS -static const struct intel_device_info intel_cannonlake_gt2_info = { +static const struct intel_device_info intel_cannonlake_info = { GEN10_FEATURES, .is_alpha_support = 1, .platform = INTEL_CANNONLAKE, @@ -579,6 +579,19 @@ static const struct intel_device_info intel_cannonlake_gt2_info = { .gt = 2, }; +#define GEN11_FEATURES \ + GEN10_FEATURES, \ + .gen = 11, \ + .ddb_size = 2048, \ + .has_csr = 0 + +static const struct intel_device_info intel_icelake_11_info = { + GEN11_FEATURES, + .platform = INTEL_ICELAKE, + .is_alpha_support = 1, + .has_resource_streamer = 0, +}; + /* * Make sure any device matches here are from most specific to most * general. For example, since the Quanta match is based on the subsystem @@ -636,8 +649,7 @@ static const struct pci_device_id pciidlist[] = { INTEL_CFL_U_GT1_IDS(&intel_coffeelake_gt1_info), INTEL_CFL_U_GT2_IDS(&intel_coffeelake_gt2_info), INTEL_CFL_U_GT3_IDS(&intel_coffeelake_gt3_info), - INTEL_CNL_U_GT2_IDS(&intel_cannonlake_gt2_info), - INTEL_CNL_Y_GT2_IDS(&intel_cannonlake_gt2_info), + INTEL_CNL_IDS(&intel_cannonlake_info), {0, 0, 0} }; MODULE_DEVICE_TABLE(pci, pciidlist); diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c index 55a8a1e29424..1c440460255d 100644 --- a/drivers/gpu/drm/i915/i915_pmu.c +++ b/drivers/gpu/drm/i915/i915_pmu.c @@ -285,28 +285,69 @@ static u64 count_interrupts(struct drm_i915_private *i915) return sum; } -static void i915_pmu_event_destroy(struct perf_event *event) +static void engine_event_destroy(struct perf_event *event) { - WARN_ON(event->parent); + struct drm_i915_private *i915 = + container_of(event->pmu, typeof(*i915), pmu.base); + struct intel_engine_cs *engine; + + engine = intel_engine_lookup_user(i915, + engine_event_class(event), + engine_event_instance(event)); + if (WARN_ON_ONCE(!engine)) + return; + + if (engine_event_sample(event) == I915_SAMPLE_BUSY && + intel_engine_supports_stats(engine)) + intel_disable_engine_stats(engine); } -static int engine_event_init(struct perf_event *event) +static void i915_pmu_event_destroy(struct perf_event *event) { - struct drm_i915_private *i915 = - container_of(event->pmu, typeof(*i915), pmu.base); + WARN_ON(event->parent); - if (!intel_engine_lookup_user(i915, engine_event_class(event), - engine_event_instance(event))) - return -ENODEV; + if (is_engine_event(event)) + engine_event_destroy(event); +} - switch (engine_event_sample(event)) { +static int +engine_event_status(struct intel_engine_cs *engine, + enum drm_i915_pmu_engine_sample sample) +{ + switch (sample) { case I915_SAMPLE_BUSY: case I915_SAMPLE_WAIT: break; case I915_SAMPLE_SEMA: + if (INTEL_GEN(engine->i915) < 6) + return -ENODEV; + break; + default: + return -ENOENT; + } + + return 0; +} + +static int +config_status(struct drm_i915_private *i915, u64 config) +{ + switch (config) { + case I915_PMU_ACTUAL_FREQUENCY: + if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915)) + /* Requires a mutex for sampling! */ + return -ENODEV; + /* Fall-through. */ + case I915_PMU_REQUESTED_FREQUENCY: if (INTEL_GEN(i915) < 6) return -ENODEV; break; + case I915_PMU_INTERRUPTS: + break; + case I915_PMU_RC6_RESIDENCY: + if (!HAS_RC6(i915)) + return -ENODEV; + break; default: return -ENOENT; } @@ -314,6 +355,30 @@ static int engine_event_init(struct perf_event *event) return 0; } +static int engine_event_init(struct perf_event *event) +{ + struct drm_i915_private *i915 = + container_of(event->pmu, typeof(*i915), pmu.base); + struct intel_engine_cs *engine; + u8 sample; + int ret; + + engine = intel_engine_lookup_user(i915, engine_event_class(event), + engine_event_instance(event)); + if (!engine) + return -ENODEV; + + sample = engine_event_sample(event); + ret = engine_event_status(engine, sample); + if (ret) + return ret; + + if (sample == I915_SAMPLE_BUSY && intel_engine_supports_stats(engine)) + ret = intel_enable_engine_stats(engine); + + return ret; +} + static int i915_pmu_event_init(struct perf_event *event) { struct drm_i915_private *i915 = @@ -337,30 +402,10 @@ static int i915_pmu_event_init(struct perf_event *event) if (!cpumask_test_cpu(event->cpu, &i915_pmu_cpumask)) return -EINVAL; - if (is_engine_event(event)) { + if (is_engine_event(event)) ret = engine_event_init(event); - } else { - ret = 0; - switch (event->attr.config) { - case I915_PMU_ACTUAL_FREQUENCY: - if (IS_VALLEYVIEW(i915) || IS_CHERRYVIEW(i915)) - /* Requires a mutex for sampling! */ - ret = -ENODEV; - case I915_PMU_REQUESTED_FREQUENCY: - if (INTEL_GEN(i915) < 6) - ret = -ENODEV; - break; - case I915_PMU_INTERRUPTS: - break; - case I915_PMU_RC6_RESIDENCY: - if (!HAS_RC6(i915)) - ret = -ENODEV; - break; - default: - ret = -ENOENT; - break; - } - } + else + ret = config_status(i915, event->attr.config); if (ret) return ret; @@ -387,7 +432,7 @@ static u64 __i915_pmu_event_read(struct perf_event *event) if (WARN_ON_ONCE(!engine)) { /* Do nothing */ } else if (sample == I915_SAMPLE_BUSY && - engine->pmu.busy_stats) { + intel_engine_supports_stats(engine)) { val = ktime_to_ns(intel_engine_get_busy_time(engine)); } else { val = engine->pmu.sample[sample].cur; @@ -442,12 +487,6 @@ again: local64_add(new - prev, &event->count); } -static bool engine_needs_busy_stats(struct intel_engine_cs *engine) -{ - return intel_engine_supports_stats(engine) && - (engine->pmu.enable & BIT(I915_SAMPLE_BUSY)); -} - static void i915_pmu_enable(struct perf_event *event) { struct drm_i915_private *i915 = @@ -487,21 +526,7 @@ static void i915_pmu_enable(struct perf_event *event) GEM_BUG_ON(sample >= I915_PMU_SAMPLE_BITS); GEM_BUG_ON(engine->pmu.enable_count[sample] == ~0); - if (engine->pmu.enable_count[sample]++ == 0) { - /* - * Enable engine busy stats tracking if needed or - * alternatively cancel the scheduled disable. - * - * If the delayed disable was pending, cancel it and - * in this case do not enable since it already is. - */ - if (engine_needs_busy_stats(engine) && - !engine->pmu.busy_stats) { - engine->pmu.busy_stats = true; - if (!cancel_delayed_work(&engine->pmu.disable_busy_stats)) - intel_enable_engine_stats(engine); - } - } + engine->pmu.enable_count[sample]++; } /* @@ -514,14 +539,6 @@ static void i915_pmu_enable(struct perf_event *event) spin_unlock_irqrestore(&i915->pmu.lock, flags); } -static void __disable_busy_stats(struct work_struct *work) -{ - struct intel_engine_cs *engine = - container_of(work, typeof(*engine), pmu.disable_busy_stats.work); - - intel_disable_engine_stats(engine); -} - static void i915_pmu_disable(struct perf_event *event) { struct drm_i915_private *i915 = @@ -545,26 +562,8 @@ static void i915_pmu_disable(struct perf_event *event) * Decrement the reference count and clear the enabled * bitmask when the last listener on an event goes away. */ - if (--engine->pmu.enable_count[sample] == 0) { + if (--engine->pmu.enable_count[sample] == 0) engine->pmu.enable &= ~BIT(sample); - if (!engine_needs_busy_stats(engine) && - engine->pmu.busy_stats) { - engine->pmu.busy_stats = false; - /* - * We request a delayed disable to handle the - * rapid on/off cycles on events, which can - * happen when tools like perf stat start, in a - * nicer way. - * - * In addition, this also helps with busy stats - * accuracy with background CPU offline/online - * migration events. - */ - queue_delayed_work(system_wq, - &engine->pmu.disable_busy_stats, - round_jiffies_up_relative(HZ)); - } - } } GEM_BUG_ON(bit >= I915_PMU_MASK_BITS); @@ -657,52 +656,9 @@ static ssize_t i915_pmu_event_show(struct device *dev, return sprintf(buf, "config=0x%lx\n", eattr->val); } -#define I915_EVENT_ATTR(_name, _config) \ - (&((struct i915_ext_attribute[]) { \ - { .attr = __ATTR(_name, 0444, i915_pmu_event_show, NULL), \ - .val = _config, } \ - })[0].attr.attr) - -#define I915_EVENT_STR(_name, _str) \ - (&((struct perf_pmu_events_attr[]) { \ - { .attr = __ATTR(_name, 0444, perf_event_sysfs_show, NULL), \ - .id = 0, \ - .event_str = _str, } \ - })[0].attr.attr) - -#define I915_EVENT(_name, _config, _unit) \ - I915_EVENT_ATTR(_name, _config), \ - I915_EVENT_STR(_name.unit, _unit) - -#define I915_ENGINE_EVENT(_name, _class, _instance, _sample) \ - I915_EVENT_ATTR(_name, __I915_PMU_ENGINE(_class, _instance, _sample)), \ - I915_EVENT_STR(_name.unit, "ns") - -#define I915_ENGINE_EVENTS(_name, _class, _instance) \ - I915_ENGINE_EVENT(_name##_instance-busy, _class, _instance, I915_SAMPLE_BUSY), \ - I915_ENGINE_EVENT(_name##_instance-sema, _class, _instance, I915_SAMPLE_SEMA), \ - I915_ENGINE_EVENT(_name##_instance-wait, _class, _instance, I915_SAMPLE_WAIT) - -static struct attribute *i915_pmu_events_attrs[] = { - I915_ENGINE_EVENTS(rcs, I915_ENGINE_CLASS_RENDER, 0), - I915_ENGINE_EVENTS(bcs, I915_ENGINE_CLASS_COPY, 0), - I915_ENGINE_EVENTS(vcs, I915_ENGINE_CLASS_VIDEO, 0), - I915_ENGINE_EVENTS(vcs, I915_ENGINE_CLASS_VIDEO, 1), - I915_ENGINE_EVENTS(vecs, I915_ENGINE_CLASS_VIDEO_ENHANCE, 0), - - I915_EVENT(actual-frequency, I915_PMU_ACTUAL_FREQUENCY, "MHz"), - I915_EVENT(requested-frequency, I915_PMU_REQUESTED_FREQUENCY, "MHz"), - - I915_EVENT_ATTR(interrupts, I915_PMU_INTERRUPTS), - - I915_EVENT(rc6-residency, I915_PMU_RC6_RESIDENCY, "ns"), - - NULL, -}; - -static const struct attribute_group i915_pmu_events_attr_group = { +static struct attribute_group i915_pmu_events_attr_group = { .name = "events", - .attrs = i915_pmu_events_attrs, + /* Patch in attrs at runtime. */ }; static ssize_t @@ -720,7 +676,7 @@ static struct attribute *i915_cpumask_attrs[] = { NULL, }; -static struct attribute_group i915_pmu_cpumask_attr_group = { +static const struct attribute_group i915_pmu_cpumask_attr_group = { .attrs = i915_cpumask_attrs, }; @@ -731,6 +687,193 @@ static const struct attribute_group *i915_pmu_attr_groups[] = { NULL }; +#define __event(__config, __name, __unit) \ +{ \ + .config = (__config), \ + .name = (__name), \ + .unit = (__unit), \ +} + +#define __engine_event(__sample, __name) \ +{ \ + .sample = (__sample), \ + .name = (__name), \ +} + +static struct i915_ext_attribute * +add_i915_attr(struct i915_ext_attribute *attr, const char *name, u64 config) +{ + sysfs_attr_init(&attr->attr.attr); + attr->attr.attr.name = name; + attr->attr.attr.mode = 0444; + attr->attr.show = i915_pmu_event_show; + attr->val = config; + + return ++attr; +} + +static struct perf_pmu_events_attr * +add_pmu_attr(struct perf_pmu_events_attr *attr, const char *name, + const char *str) +{ + sysfs_attr_init(&attr->attr.attr); + attr->attr.attr.name = name; + attr->attr.attr.mode = 0444; + attr->attr.show = perf_event_sysfs_show; + attr->event_str = str; + + return ++attr; +} + +static struct attribute ** +create_event_attributes(struct drm_i915_private *i915) +{ + static const struct { + u64 config; + const char *name; + const char *unit; + } events[] = { + __event(I915_PMU_ACTUAL_FREQUENCY, "actual-frequency", "MHz"), + __event(I915_PMU_REQUESTED_FREQUENCY, "requested-frequency", "MHz"), + __event(I915_PMU_INTERRUPTS, "interrupts", NULL), + __event(I915_PMU_RC6_RESIDENCY, "rc6-residency", "ns"), + }; + static const struct { + enum drm_i915_pmu_engine_sample sample; + char *name; + } engine_events[] = { + __engine_event(I915_SAMPLE_BUSY, "busy"), + __engine_event(I915_SAMPLE_SEMA, "sema"), + __engine_event(I915_SAMPLE_WAIT, "wait"), + }; + unsigned int count = 0; + struct perf_pmu_events_attr *pmu_attr = NULL, *pmu_iter; + struct i915_ext_attribute *i915_attr = NULL, *i915_iter; + struct attribute **attr = NULL, **attr_iter; + struct intel_engine_cs *engine; + enum intel_engine_id id; + unsigned int i; + + /* Count how many counters we will be exposing. */ + for (i = 0; i < ARRAY_SIZE(events); i++) { + if (!config_status(i915, events[i].config)) + count++; + } + + for_each_engine(engine, i915, id) { + for (i = 0; i < ARRAY_SIZE(engine_events); i++) { + if (!engine_event_status(engine, + engine_events[i].sample)) + count++; + } + } + + /* Allocate attribute objects and table. */ + i915_attr = kcalloc(count, sizeof(*i915_attr), GFP_KERNEL); + if (!i915_attr) + goto err_alloc; + + pmu_attr = kcalloc(count, sizeof(*pmu_attr), GFP_KERNEL); + if (!pmu_attr) + goto err_alloc; + + /* Max one pointer of each attribute type plus a termination entry. */ + attr = kcalloc(count * 2 + 1, sizeof(*attr), GFP_KERNEL); + if (!attr) + goto err_alloc; + + i915_iter = i915_attr; + pmu_iter = pmu_attr; + attr_iter = attr; + + /* Initialize supported non-engine counters. */ + for (i = 0; i < ARRAY_SIZE(events); i++) { + char *str; + + if (config_status(i915, events[i].config)) + continue; + + str = kstrdup(events[i].name, GFP_KERNEL); + if (!str) + goto err; + + *attr_iter++ = &i915_iter->attr.attr; + i915_iter = add_i915_attr(i915_iter, str, events[i].config); + + if (events[i].unit) { + str = kasprintf(GFP_KERNEL, "%s.unit", events[i].name); + if (!str) + goto err; + + *attr_iter++ = &pmu_iter->attr.attr; + pmu_iter = add_pmu_attr(pmu_iter, str, events[i].unit); + } + } + + /* Initialize supported engine counters. */ + for_each_engine(engine, i915, id) { + for (i = 0; i < ARRAY_SIZE(engine_events); i++) { + char *str; + + if (engine_event_status(engine, + engine_events[i].sample)) + continue; + + str = kasprintf(GFP_KERNEL, "%s-%s", + engine->name, engine_events[i].name); + if (!str) + goto err; + + *attr_iter++ = &i915_iter->attr.attr; + i915_iter = + add_i915_attr(i915_iter, str, + __I915_PMU_ENGINE(engine->uabi_class, + engine->instance, + engine_events[i].sample)); + + str = kasprintf(GFP_KERNEL, "%s-%s.unit", + engine->name, engine_events[i].name); + if (!str) + goto err; + + *attr_iter++ = &pmu_iter->attr.attr; + pmu_iter = add_pmu_attr(pmu_iter, str, "ns"); + } + } + + i915->pmu.i915_attr = i915_attr; + i915->pmu.pmu_attr = pmu_attr; + + return attr; + +err:; + for (attr_iter = attr; *attr_iter; attr_iter++) + kfree((*attr_iter)->name); + +err_alloc: + kfree(attr); + kfree(i915_attr); + kfree(pmu_attr); + + return NULL; +} + +static void free_event_attributes(struct drm_i915_private *i915) +{ + struct attribute **attr_iter = i915_pmu_events_attr_group.attrs; + + for (; *attr_iter; attr_iter++) + kfree((*attr_iter)->name); + + kfree(i915_pmu_events_attr_group.attrs); + kfree(i915->pmu.i915_attr); + kfree(i915->pmu.pmu_attr); + + i915_pmu_events_attr_group.attrs = NULL; + i915->pmu.i915_attr = NULL; + i915->pmu.pmu_attr = NULL; +} + static int i915_pmu_cpu_online(unsigned int cpu, struct hlist_node *node) { struct i915_pmu *pmu = hlist_entry_safe(node, typeof(*pmu), node); @@ -797,8 +940,6 @@ static void i915_pmu_unregister_cpuhp_state(struct drm_i915_private *i915) void i915_pmu_register(struct drm_i915_private *i915) { - struct intel_engine_cs *engine; - enum intel_engine_id id; int ret; if (INTEL_GEN(i915) <= 2) { @@ -806,6 +947,12 @@ void i915_pmu_register(struct drm_i915_private *i915) return; } + i915_pmu_events_attr_group.attrs = create_event_attributes(i915); + if (!i915_pmu_events_attr_group.attrs) { + ret = -ENOMEM; + goto err; + } + i915->pmu.base.attr_groups = i915_pmu_attr_groups; i915->pmu.base.task_ctx_nr = perf_invalid_context; i915->pmu.base.event_init = i915_pmu_event_init; @@ -820,10 +967,6 @@ void i915_pmu_register(struct drm_i915_private *i915) hrtimer_init(&i915->pmu.timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); i915->pmu.timer.function = i915_sample; - for_each_engine(engine, i915, id) - INIT_DELAYED_WORK(&engine->pmu.disable_busy_stats, - __disable_busy_stats); - ret = perf_pmu_register(&i915->pmu.base, "i915", -1); if (ret) goto err; @@ -838,14 +981,12 @@ err_unreg: perf_pmu_unregister(&i915->pmu.base); err: i915->pmu.base.event_init = NULL; + free_event_attributes(i915); DRM_NOTE("Failed to register PMU! (err=%d)\n", ret); } void i915_pmu_unregister(struct drm_i915_private *i915) { - struct intel_engine_cs *engine; - enum intel_engine_id id; - if (!i915->pmu.base.event_init) return; @@ -853,13 +994,9 @@ void i915_pmu_unregister(struct drm_i915_private *i915) hrtimer_cancel(&i915->pmu.timer); - for_each_engine(engine, i915, id) { - GEM_BUG_ON(engine->pmu.busy_stats); - flush_delayed_work(&engine->pmu.disable_busy_stats); - } - i915_pmu_unregister_cpuhp_state(i915); perf_pmu_unregister(&i915->pmu.base); i915->pmu.base.event_init = NULL; + free_event_attributes(i915); } diff --git a/drivers/gpu/drm/i915/i915_pmu.h b/drivers/gpu/drm/i915/i915_pmu.h index 40c154d13565..5a2e013a56bb 100644 --- a/drivers/gpu/drm/i915/i915_pmu.h +++ b/drivers/gpu/drm/i915/i915_pmu.h @@ -94,6 +94,14 @@ struct i915_pmu { * struct intel_engine_cs. */ struct i915_pmu_sample sample[__I915_NUM_PMU_SAMPLERS]; + /** + * @i915_attr: Memory block holding device attributes. + */ + void *i915_attr; + /** + * @pmu_attr: Memory block holding device attributes. + */ + void *pmu_attr; }; #ifdef CONFIG_PERF_EVENTS diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index a2108e35c599..1489dd3b3ec2 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -1304,6 +1304,7 @@ enum i915_power_well_id { SKL_DISP_PW_DDI_B, SKL_DISP_PW_DDI_C, SKL_DISP_PW_DDI_D, + CNL_DISP_PW_DDI_F = 6, GLK_DISP_PW_AUX_A = 8, GLK_DISP_PW_AUX_B, @@ -1312,6 +1313,7 @@ enum i915_power_well_id { CNL_DISP_PW_AUX_B = GLK_DISP_PW_AUX_B, CNL_DISP_PW_AUX_C = GLK_DISP_PW_AUX_C, CNL_DISP_PW_AUX_D, + CNL_DISP_PW_AUX_F, SKL_DISP_PW_1 = 14, SKL_DISP_PW_2, @@ -1963,7 +1965,7 @@ enum i915_power_well_id { #define _CNL_PORT_TX_DW2_LN0_B 0x162648 #define _CNL_PORT_TX_DW2_LN0_C 0x162C48 #define _CNL_PORT_TX_DW2_LN0_D 0x162E48 -#define _CNL_PORT_TX_DW2_LN0_F 0x162A48 +#define _CNL_PORT_TX_DW2_LN0_F 0x162848 #define CNL_PORT_TX_DW2_GRP(port) _MMIO_PORT6(port, \ _CNL_PORT_TX_DW2_GRP_AE, \ _CNL_PORT_TX_DW2_GRP_B, \ @@ -2590,6 +2592,8 @@ enum i915_power_well_id { #define GFX_FORWARD_VBLANK_ALWAYS (1<<5) #define GFX_FORWARD_VBLANK_COND (2<<5) +#define GEN11_GFX_DISABLE_LEGACY_MODE (1<<3) + #define VLV_DISPLAY_BASE 0x180000 #define VLV_MIPI_BASE VLV_DISPLAY_BASE #define BXT_MIPI_BASE 0x60000 @@ -2648,6 +2652,31 @@ enum i915_power_well_id { #define LM_FIFO_WATERMARK 0x0000001F #define MI_ARB_STATE _MMIO(0x20e4) /* 915+ only */ +#define MBUS_ABOX_CTL _MMIO(0x45038) +#define MBUS_ABOX_BW_CREDIT_MASK (3 << 20) +#define MBUS_ABOX_BW_CREDIT(x) ((x) << 20) +#define MBUS_ABOX_B_CREDIT_MASK (0xF << 16) +#define MBUS_ABOX_B_CREDIT(x) ((x) << 16) +#define MBUS_ABOX_BT_CREDIT_POOL2_MASK (0x1F << 8) +#define MBUS_ABOX_BT_CREDIT_POOL2(x) ((x) << 8) +#define MBUS_ABOX_BT_CREDIT_POOL1_MASK (0x1F << 0) +#define MBUS_ABOX_BT_CREDIT_POOL1(x) ((x) << 0) + +#define _PIPEA_MBUS_DBOX_CTL 0x7003C +#define _PIPEB_MBUS_DBOX_CTL 0x7103C +#define PIPE_MBUS_DBOX_CTL(pipe) _MMIO_PIPE(pipe, _PIPEA_MBUS_DBOX_CTL, \ + _PIPEB_MBUS_DBOX_CTL) +#define MBUS_DBOX_BW_CREDIT_MASK (3 << 14) +#define MBUS_DBOX_BW_CREDIT(x) ((x) << 14) +#define MBUS_DBOX_B_CREDIT_MASK (0x1F << 8) +#define MBUS_DBOX_B_CREDIT(x) ((x) << 8) +#define MBUS_DBOX_A_CREDIT_MASK (0xF << 0) +#define MBUS_DBOX_A_CREDIT(x) ((x) << 0) + +#define MBUS_UBOX_CTL _MMIO(0x4503C) +#define MBUS_BBOX_CTL_S1 _MMIO(0x45040) +#define MBUS_BBOX_CTL_S2 _MMIO(0x45044) + /* Make render/texture TLB fetches lower priorty than associated data * fetches. This is not turned on by default */ @@ -3063,7 +3092,12 @@ enum i915_power_well_id { #define GMBUS_PIN_2_BXT 2 #define GMBUS_PIN_3_BXT 3 #define GMBUS_PIN_4_CNP 4 -#define GMBUS_NUM_PINS 7 /* including 0 */ +#define GMBUS_PIN_9_TC1_ICP 9 +#define GMBUS_PIN_10_TC2_ICP 10 +#define GMBUS_PIN_11_TC3_ICP 11 +#define GMBUS_PIN_12_TC4_ICP 12 + +#define GMBUS_NUM_PINS 13 /* including 0 */ #define GMBUS1 _MMIO(dev_priv->gpio_mmio_base + 0x5104) /* command/status */ #define GMBUS_SW_CLR_INT (1<<31) #define GMBUS_SW_RDY (1<<30) @@ -4066,7 +4100,7 @@ enum { #define EDP_PSR_AUX_CTL _MMIO(dev_priv->psr_mmio_base + 0x10) #define EDP_PSR_AUX_DATA(i) _MMIO(dev_priv->psr_mmio_base + 0x14 + (i) * 4) /* 5 registers */ -#define EDP_PSR_STATUS_CTL _MMIO(dev_priv->psr_mmio_base + 0x40) +#define EDP_PSR_STATUS _MMIO(dev_priv->psr_mmio_base + 0x40) #define EDP_PSR_STATUS_STATE_MASK (7<<29) #define EDP_PSR_STATUS_STATE_IDLE (0<<29) #define EDP_PSR_STATUS_STATE_SRDONACK (1<<29) @@ -4093,7 +4127,7 @@ enum { #define EDP_PSR_PERF_CNT _MMIO(dev_priv->psr_mmio_base + 0x44) #define EDP_PSR_PERF_CNT_MASK 0xffffff -#define EDP_PSR_DEBUG_CTL _MMIO(dev_priv->psr_mmio_base + 0x60) +#define EDP_PSR_DEBUG _MMIO(dev_priv->psr_mmio_base + 0x60) #define EDP_PSR_DEBUG_MASK_MAX_SLEEP (1<<28) #define EDP_PSR_DEBUG_MASK_LPSP (1<<27) #define EDP_PSR_DEBUG_MASK_MEMUP (1<<26) @@ -4116,7 +4150,7 @@ enum { #define EDP_PSR2_IDLE_MASK 0xf #define EDP_PSR2_FRAME_BEFORE_SU(a) ((a)<<4) -#define EDP_PSR2_STATUS_CTL _MMIO(0x6f940) +#define EDP_PSR2_STATUS _MMIO(0x6f940) #define EDP_PSR2_STATUS_STATE_MASK (0xf<<28) #define EDP_PSR2_STATUS_STATE_SHIFT 28 @@ -5278,6 +5312,13 @@ enum { #define _DPD_AUX_CH_DATA4 (dev_priv->info.display_mmio_offset + 0x64320) #define _DPD_AUX_CH_DATA5 (dev_priv->info.display_mmio_offset + 0x64324) +#define _DPF_AUX_CH_CTL (dev_priv->info.display_mmio_offset + 0x64510) +#define _DPF_AUX_CH_DATA1 (dev_priv->info.display_mmio_offset + 0x64514) +#define _DPF_AUX_CH_DATA2 (dev_priv->info.display_mmio_offset + 0x64518) +#define _DPF_AUX_CH_DATA3 (dev_priv->info.display_mmio_offset + 0x6451c) +#define _DPF_AUX_CH_DATA4 (dev_priv->info.display_mmio_offset + 0x64520) +#define _DPF_AUX_CH_DATA5 (dev_priv->info.display_mmio_offset + 0x64524) + #define DP_AUX_CH_CTL(port) _MMIO_PORT(port, _DPA_AUX_CH_CTL, _DPB_AUX_CH_CTL) #define DP_AUX_CH_DATA(port, i) _MMIO(_PORT(port, _DPA_AUX_CH_DATA1, _DPB_AUX_CH_DATA1) + (i) * 4) /* 5 registers */ @@ -6310,6 +6351,11 @@ enum { #define _PLANE_CTL_3_A 0x70380 #define PLANE_CTL_ENABLE (1 << 31) #define PLANE_CTL_PIPE_GAMMA_ENABLE (1 << 30) /* Pre-GLK */ +/* + * ICL+ uses the same PLANE_CTL_FORMAT bits, but the field definition + * expanded to include bit 23 as well. However, the shift-24 based values + * correctly map to the same formats in ICL, as long as bit 23 is set to 0 + */ #define PLANE_CTL_FORMAT_MASK (0xf << 24) #define PLANE_CTL_FORMAT_YUV422 ( 0 << 24) #define PLANE_CTL_FORMAT_NV12 ( 1 << 24) @@ -6319,6 +6365,7 @@ enum { #define PLANE_CTL_FORMAT_AYUV ( 8 << 24) #define PLANE_CTL_FORMAT_INDEXED ( 12 << 24) #define PLANE_CTL_FORMAT_RGB_565 ( 14 << 24) +#define ICL_PLANE_CTL_FORMAT_MASK (0x1f << 23) #define PLANE_CTL_PIPE_CSC_ENABLE (1 << 23) /* Pre-GLK */ #define PLANE_CTL_KEY_ENABLE_MASK (0x3 << 21) #define PLANE_CTL_KEY_ENABLE_SOURCE ( 1 << 21) @@ -6933,6 +6980,7 @@ enum { #define GEN8_DE_PORT_IMR _MMIO(0x44444) #define GEN8_DE_PORT_IIR _MMIO(0x44448) #define GEN8_DE_PORT_IER _MMIO(0x4444c) +#define CNL_AUX_CHANNEL_F (1 << 28) #define GEN9_AUX_CHANNEL_D (1 << 27) #define GEN9_AUX_CHANNEL_C (1 << 26) #define GEN9_AUX_CHANNEL_B (1 << 25) @@ -6957,6 +7005,69 @@ enum { #define GEN8_PCU_IIR _MMIO(0x444e8) #define GEN8_PCU_IER _MMIO(0x444ec) +#define GEN11_GFX_MSTR_IRQ _MMIO(0x190010) +#define GEN11_MASTER_IRQ (1 << 31) +#define GEN11_PCU_IRQ (1 << 30) +#define GEN11_DISPLAY_IRQ (1 << 16) +#define GEN11_GT_DW_IRQ(x) (1 << (x)) +#define GEN11_GT_DW1_IRQ (1 << 1) +#define GEN11_GT_DW0_IRQ (1 << 0) + +#define GEN11_DISPLAY_INT_CTL _MMIO(0x44200) +#define GEN11_DISPLAY_IRQ_ENABLE (1 << 31) +#define GEN11_AUDIO_CODEC_IRQ (1 << 24) +#define GEN11_DE_PCH_IRQ (1 << 23) +#define GEN11_DE_MISC_IRQ (1 << 22) +#define GEN11_DE_PORT_IRQ (1 << 20) +#define GEN11_DE_PIPE_C (1 << 18) +#define GEN11_DE_PIPE_B (1 << 17) +#define GEN11_DE_PIPE_A (1 << 16) + +#define GEN11_GT_INTR_DW0 _MMIO(0x190018) +#define GEN11_CSME (31) +#define GEN11_GUNIT (28) +#define GEN11_GUC (25) +#define GEN11_WDPERF (20) +#define GEN11_KCR (19) +#define GEN11_GTPM (16) +#define GEN11_BCS (15) +#define GEN11_RCS0 (0) + +#define GEN11_GT_INTR_DW1 _MMIO(0x19001c) +#define GEN11_VECS(x) (31 - (x)) +#define GEN11_VCS(x) (x) + +#define GEN11_GT_INTR_DW(x) _MMIO(0x190018 + (x * 4)) + +#define GEN11_INTR_IDENTITY_REG0 _MMIO(0x190060) +#define GEN11_INTR_IDENTITY_REG1 _MMIO(0x190064) +#define GEN11_INTR_DATA_VALID (1 << 31) +#define GEN11_INTR_ENGINE_MASK (0xffff) + +#define GEN11_INTR_IDENTITY_REG(x) _MMIO(0x190060 + (x * 4)) + +#define GEN11_IIR_REG0_SELECTOR _MMIO(0x190070) +#define GEN11_IIR_REG1_SELECTOR _MMIO(0x190074) + +#define GEN11_IIR_REG_SELECTOR(x) _MMIO(0x190070 + (x * 4)) + +#define GEN11_RENDER_COPY_INTR_ENABLE _MMIO(0x190030) +#define GEN11_VCS_VECS_INTR_ENABLE _MMIO(0x190034) +#define GEN11_GUC_SG_INTR_ENABLE _MMIO(0x190038) +#define GEN11_GPM_WGBOXPERF_INTR_ENABLE _MMIO(0x19003c) +#define GEN11_CRYPTO_RSVD_INTR_ENABLE _MMIO(0x190040) +#define GEN11_GUNIT_CSME_INTR_ENABLE _MMIO(0x190044) + +#define GEN11_RCS0_RSVD_INTR_MASK _MMIO(0x190090) +#define GEN11_BCS_RSVD_INTR_MASK _MMIO(0x1900a0) +#define GEN11_VCS0_VCS1_INTR_MASK _MMIO(0x1900a8) +#define GEN11_VCS2_VCS3_INTR_MASK _MMIO(0x1900ac) +#define GEN11_VECS0_VECS1_INTR_MASK _MMIO(0x1900d0) +#define GEN11_GUC_SG_INTR_MASK _MMIO(0x1900e8) +#define GEN11_GPM_WGBOXPERF_INTR_MASK _MMIO(0x1900ec) +#define GEN11_CRYPTO_RSVD_INTR_MASK _MMIO(0x1900f0) +#define GEN11_GUNIT_CSME_INTR_MASK _MMIO(0x1900f4) + #define ILK_DISPLAY_CHICKEN2 _MMIO(0x42004) /* Required on all Ironlake and Sandybridge according to the B-Spec. */ #define ILK_ELPIN_409_SELECT (1 << 25) @@ -7011,8 +7122,12 @@ enum { #define CHICKEN_TRANS_A 0x420c0 #define CHICKEN_TRANS_B 0x420c4 #define CHICKEN_TRANS(trans) _MMIO_TRANS(trans, CHICKEN_TRANS_A, CHICKEN_TRANS_B) -#define PSR2_VSC_ENABLE_PROG_HEADER (1<<12) -#define PSR2_ADD_VERTICAL_LINE_COUNT (1<<15) +#define DDI_TRAINING_OVERRIDE_ENABLE (1<<19) +#define DDI_TRAINING_OVERRIDE_VALUE (1<<18) +#define DDIE_TRAINING_OVERRIDE_ENABLE (1<<17) /* CHICKEN_TRANS_A only */ +#define DDIE_TRAINING_OVERRIDE_VALUE (1<<16) /* CHICKEN_TRANS_A only */ +#define PSR2_ADD_VERTICAL_LINE_COUNT (1<<15) +#define PSR2_VSC_ENABLE_PROG_HEADER (1<<12) #define DISP_ARB_CTL _MMIO(0x45000) #define DISP_FBC_MEMORY_WAKE (1<<31) @@ -7351,6 +7466,8 @@ enum { #define CNP_RAWCLK_DIV(div) ((div) << 16) #define CNP_RAWCLK_FRAC_MASK (0xf << 26) #define CNP_RAWCLK_FRAC(frac) ((frac) << 26) +#define ICP_RAWCLK_DEN(den) ((den) << 26) +#define ICP_RAWCLK_NUM(num) ((num) << 11) #define PCH_DPLL_TMR_CFG _MMIO(0xc6208) @@ -8350,6 +8467,19 @@ enum skl_power_gate { #define SKL_PW_TO_PG(pw) ((pw) - SKL_DISP_PW_1 + SKL_PG1) #define SKL_FUSE_PG_DIST_STATUS(pg) (1 << (27 - (pg))) +#define _CNL_AUX_REG_IDX(pw) ((pw) - 9) +#define _CNL_AUX_ANAOVRD1_B 0x162250 +#define _CNL_AUX_ANAOVRD1_C 0x162210 +#define _CNL_AUX_ANAOVRD1_D 0x1622D0 +#define _CNL_AUX_ANAOVRD1_F 0x162A90 +#define CNL_AUX_ANAOVRD1(pw) _MMIO(_PICK(_CNL_AUX_REG_IDX(pw), \ + _CNL_AUX_ANAOVRD1_B, \ + _CNL_AUX_ANAOVRD1_C, \ + _CNL_AUX_ANAOVRD1_D, \ + _CNL_AUX_ANAOVRD1_F)) +#define CNL_AUX_ANAOVRD1_ENABLE (1<<16) +#define CNL_AUX_ANAOVRD1_LDO_BYPASS (1<<23) + /* Per-pipe DDI Function Control */ #define _TRANS_DDI_FUNC_CTL_A 0x60400 #define _TRANS_DDI_FUNC_CTL_B 0x61400 @@ -8663,10 +8793,12 @@ enum skl_power_gate { * CNL Clocks */ #define DPCLKA_CFGCR0 _MMIO(0x6C200) -#define DPCLKA_CFGCR0_DDI_CLK_OFF(port) (1 << ((port)+10)) -#define DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(port) (3 << ((port)*2)) -#define DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(port) ((port)*2) -#define DPCLKA_CFGCR0_DDI_CLK_SEL(pll, port) ((pll) << ((port)*2)) +#define DPCLKA_CFGCR0_DDI_CLK_OFF(port) (1 << ((port) == PORT_F ? 23 : \ + (port)+10)) +#define DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(port) ((port) == PORT_F ? 21 : \ + (port)*2) +#define DPCLKA_CFGCR0_DDI_CLK_SEL_MASK(port) (3 << DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(port)) +#define DPCLKA_CFGCR0_DDI_CLK_SEL(pll, port) ((pll) << DPCLKA_CFGCR0_DDI_CLK_SEL_SHIFT(port)) /* CNL PLL */ #define DPLL0_ENABLE 0x46010 @@ -8762,6 +8894,7 @@ enum skl_power_gate { #define SFUSE_STRAP_RAW_FREQUENCY (1<<8) #define SFUSE_STRAP_DISPLAY_DISABLED (1<<7) #define SFUSE_STRAP_CRT_DISABLED (1<<6) +#define SFUSE_STRAP_DDIF_DETECTED (1<<3) #define SFUSE_STRAP_DDIB_DETECTED (1<<2) #define SFUSE_STRAP_DDIC_DETECTED (1<<1) #define SFUSE_STRAP_DDID_DETECTED (1<<0) diff --git a/drivers/gpu/drm/i915/i915_sw_fence.c b/drivers/gpu/drm/i915/i915_sw_fence.c index 3669f5eeb91e..1de5173e53a2 100644 --- a/drivers/gpu/drm/i915/i915_sw_fence.c +++ b/drivers/gpu/drm/i915/i915_sw_fence.c @@ -365,18 +365,31 @@ int i915_sw_fence_await_sw_fence_gfp(struct i915_sw_fence *fence, struct i915_sw_dma_fence_cb { struct dma_fence_cb base; struct i915_sw_fence *fence; +}; + +struct i915_sw_dma_fence_cb_timer { + struct i915_sw_dma_fence_cb base; struct dma_fence *dma; struct timer_list timer; struct irq_work work; struct rcu_head rcu; }; +static void dma_i915_sw_fence_wake(struct dma_fence *dma, + struct dma_fence_cb *data) +{ + struct i915_sw_dma_fence_cb *cb = container_of(data, typeof(*cb), base); + + i915_sw_fence_complete(cb->fence); + kfree(cb); +} + static void timer_i915_sw_fence_wake(struct timer_list *t) { - struct i915_sw_dma_fence_cb *cb = from_timer(cb, t, timer); + struct i915_sw_dma_fence_cb_timer *cb = from_timer(cb, t, timer); struct i915_sw_fence *fence; - fence = xchg(&cb->fence, NULL); + fence = xchg(&cb->base.fence, NULL); if (!fence) return; @@ -388,13 +401,14 @@ static void timer_i915_sw_fence_wake(struct timer_list *t) i915_sw_fence_complete(fence); } -static void dma_i915_sw_fence_wake(struct dma_fence *dma, - struct dma_fence_cb *data) +static void dma_i915_sw_fence_wake_timer(struct dma_fence *dma, + struct dma_fence_cb *data) { - struct i915_sw_dma_fence_cb *cb = container_of(data, typeof(*cb), base); + struct i915_sw_dma_fence_cb_timer *cb = + container_of(data, typeof(*cb), base.base); struct i915_sw_fence *fence; - fence = xchg(&cb->fence, NULL); + fence = xchg(&cb->base.fence, NULL); if (fence) i915_sw_fence_complete(fence); @@ -403,7 +417,8 @@ static void dma_i915_sw_fence_wake(struct dma_fence *dma, static void irq_i915_sw_fence_work(struct irq_work *wrk) { - struct i915_sw_dma_fence_cb *cb = container_of(wrk, typeof(*cb), work); + struct i915_sw_dma_fence_cb_timer *cb = + container_of(wrk, typeof(*cb), work); del_timer_sync(&cb->timer); dma_fence_put(cb->dma); @@ -417,6 +432,7 @@ int i915_sw_fence_await_dma_fence(struct i915_sw_fence *fence, gfp_t gfp) { struct i915_sw_dma_fence_cb *cb; + dma_fence_func_t func; int ret; debug_fence_assert(fence); @@ -425,7 +441,10 @@ int i915_sw_fence_await_dma_fence(struct i915_sw_fence *fence, if (dma_fence_is_signaled(dma)) return 0; - cb = kmalloc(sizeof(*cb), gfp); + cb = kmalloc(timeout ? + sizeof(struct i915_sw_dma_fence_cb_timer) : + sizeof(struct i915_sw_dma_fence_cb), + gfp); if (!cb) { if (!gfpflags_allow_blocking(gfp)) return -ENOMEM; @@ -436,19 +455,26 @@ int i915_sw_fence_await_dma_fence(struct i915_sw_fence *fence, cb->fence = fence; i915_sw_fence_await(fence); - cb->dma = NULL; - timer_setup(&cb->timer, timer_i915_sw_fence_wake, TIMER_IRQSAFE); - init_irq_work(&cb->work, irq_i915_sw_fence_work); + func = dma_i915_sw_fence_wake; if (timeout) { - cb->dma = dma_fence_get(dma); - mod_timer(&cb->timer, round_jiffies_up(jiffies + timeout)); + struct i915_sw_dma_fence_cb_timer *timer = + container_of(cb, typeof(*timer), base); + + timer->dma = dma_fence_get(dma); + init_irq_work(&timer->work, irq_i915_sw_fence_work); + + timer_setup(&timer->timer, + timer_i915_sw_fence_wake, TIMER_IRQSAFE); + mod_timer(&timer->timer, round_jiffies_up(jiffies + timeout)); + + func = dma_i915_sw_fence_wake_timer; } - ret = dma_fence_add_callback(dma, &cb->base, dma_i915_sw_fence_wake); + ret = dma_fence_add_callback(dma, &cb->base, func); if (ret == 0) { ret = 1; } else { - dma_i915_sw_fence_wake(dma, &cb->base); + func(dma, &cb->base); if (ret == -ENOENT) /* fence already signaled */ ret = 0; } diff --git a/drivers/gpu/drm/i915/intel_atomic_plane.c b/drivers/gpu/drm/i915/intel_atomic_plane.c index 8e6dc159f64d..57ee8b786cd8 100644 --- a/drivers/gpu/drm/i915/intel_atomic_plane.c +++ b/drivers/gpu/drm/i915/intel_atomic_plane.c @@ -56,7 +56,6 @@ intel_create_plane_state(struct drm_plane *plane) state->base.plane = plane; state->base.rotation = DRM_MODE_ROTATE_0; - state->ckey.flags = I915_SET_COLORKEY_NONE; return state; } diff --git a/drivers/gpu/drm/i915/intel_bios.c b/drivers/gpu/drm/i915/intel_bios.c index f7f771749e48..4e74aa2f16bc 100644 --- a/drivers/gpu/drm/i915/intel_bios.c +++ b/drivers/gpu/drm/i915/intel_bios.c @@ -1146,6 +1146,7 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port, {DVO_PORT_HDMIC, DVO_PORT_DPC, -1}, {DVO_PORT_HDMID, DVO_PORT_DPD, -1}, {DVO_PORT_CRT, DVO_PORT_HDMIE, DVO_PORT_DPE}, + {DVO_PORT_HDMIF, DVO_PORT_DPF, -1}, }; /* @@ -1273,6 +1274,27 @@ static void parse_ddi_port(struct drm_i915_private *dev_priv, enum port port, DRM_DEBUG_KMS("VBT HDMI boost level for port %c: %d\n", port_name(port), info->hdmi_boost_level); } + + /* DP max link rate for CNL+ */ + if (bdb_version >= 216) { + switch (child->dp_max_link_rate) { + default: + case VBT_DP_MAX_LINK_RATE_HBR3: + info->dp_max_link_rate = 810000; + break; + case VBT_DP_MAX_LINK_RATE_HBR2: + info->dp_max_link_rate = 540000; + break; + case VBT_DP_MAX_LINK_RATE_HBR: + info->dp_max_link_rate = 270000; + break; + case VBT_DP_MAX_LINK_RATE_LBR: + info->dp_max_link_rate = 162000; + break; + } + DRM_DEBUG_KMS("VBT DP max link rate for port %c: %d\n", + port_name(port), info->dp_max_link_rate); + } } static void parse_ddi_ports(struct drm_i915_private *dev_priv, u8 bdb_version) @@ -1696,6 +1718,7 @@ bool intel_bios_is_port_present(struct drm_i915_private *dev_priv, enum port por [PORT_C] = { DVO_PORT_DPC, DVO_PORT_HDMIC, }, [PORT_D] = { DVO_PORT_DPD, DVO_PORT_HDMID, }, [PORT_E] = { DVO_PORT_DPE, DVO_PORT_HDMIE, }, + [PORT_F] = { DVO_PORT_DPF, DVO_PORT_HDMIF, }, }; int i; @@ -1734,6 +1757,7 @@ bool intel_bios_is_port_edp(struct drm_i915_private *dev_priv, enum port port) [PORT_C] = DVO_PORT_DPC, [PORT_D] = DVO_PORT_DPD, [PORT_E] = DVO_PORT_DPE, + [PORT_F] = DVO_PORT_DPF, }; int i; @@ -1769,6 +1793,7 @@ static bool child_dev_is_dp_dual_mode(const struct child_device_config *child, [PORT_C] = { DVO_PORT_DPC, DVO_PORT_HDMIC, }, [PORT_D] = { DVO_PORT_DPD, DVO_PORT_HDMID, }, [PORT_E] = { DVO_PORT_DPE, DVO_PORT_HDMIE, }, + [PORT_F] = { DVO_PORT_DPF, DVO_PORT_HDMIF, }, }; if (port == PORT_A || port >= ARRAY_SIZE(port_mapping)) @@ -1935,6 +1960,11 @@ intel_bios_is_lspcon_present(struct drm_i915_private *dev_priv, if (port == PORT_D) return true; break; + case DVO_PORT_DPF: + case DVO_PORT_HDMIF: + if (port == PORT_F) + return true; + break; default: break; } diff --git a/drivers/gpu/drm/i915/intel_breadcrumbs.c b/drivers/gpu/drm/i915/intel_breadcrumbs.c index bd40fea16b4f..efbc627a2a25 100644 --- a/drivers/gpu/drm/i915/intel_breadcrumbs.c +++ b/drivers/gpu/drm/i915/intel_breadcrumbs.c @@ -224,7 +224,7 @@ void intel_engine_disarm_breadcrumbs(struct intel_engine_cs *engine) struct intel_wait *wait, *n; if (!b->irq_armed) - goto wakeup_signaler; + return; /* * We only disarm the irq when we are idle (all requests completed), @@ -249,14 +249,6 @@ void intel_engine_disarm_breadcrumbs(struct intel_engine_cs *engine) b->waiters = RB_ROOT; spin_unlock_irq(&b->rb_lock); - - /* - * The signaling thread may be asleep holding a reference to a request, - * that had its signaling cancelled prior to being preempted. We need - * to kick the signaler, just in case, to release any such reference. - */ -wakeup_signaler: - wake_up_process(b->signaler); } static bool use_fake_irq(const struct intel_breadcrumbs *b) @@ -385,6 +377,8 @@ static bool __intel_engine_add_wait(struct intel_engine_cs *engine, bool first, armed; u32 seqno; + GEM_BUG_ON(!wait->seqno); + /* Insert the request into the retirement ordered list * of waiters by walking the rbtree. If we are the oldest * seqno in the tree (the first to be retired), then @@ -631,6 +625,63 @@ static void signaler_set_rtpriority(void) sched_setscheduler_nocheck(current, SCHED_FIFO, ¶m); } +static void __intel_engine_remove_signal(struct intel_engine_cs *engine, + struct drm_i915_gem_request *request) +{ + struct intel_breadcrumbs *b = &engine->breadcrumbs; + + lockdep_assert_held(&b->rb_lock); + + /* + * Wake up all other completed waiters and select the + * next bottom-half for the next user interrupt. + */ + __intel_engine_remove_wait(engine, &request->signaling.wait); + + /* + * Find the next oldest signal. Note that as we have + * not been holding the lock, another client may + * have installed an even older signal than the one + * we just completed - so double check we are still + * the oldest before picking the next one. + */ + if (request->signaling.wait.seqno) { + if (request == rcu_access_pointer(b->first_signal)) { + struct rb_node *rb = rb_next(&request->signaling.node); + rcu_assign_pointer(b->first_signal, + rb ? to_signaler(rb) : NULL); + } + + rb_erase(&request->signaling.node, &b->signals); + request->signaling.wait.seqno = 0; + } +} + +static struct drm_i915_gem_request * +get_first_signal_rcu(struct intel_breadcrumbs *b) +{ + /* + * See the big warnings for i915_gem_active_get_rcu() and similarly + * for dma_fence_get_rcu_safe() that explain the intricacies involved + * here with defeating CPU/compiler speculation and enforcing + * the required memory barriers. + */ + do { + struct drm_i915_gem_request *request; + + request = rcu_dereference(b->first_signal); + if (request) + request = i915_gem_request_get_rcu(request); + + barrier(); + + if (!request || request == rcu_access_pointer(b->first_signal)) + return rcu_pointer_handoff(request); + + i915_gem_request_put(request); + } while (1); +} + static int intel_breadcrumbs_signaler(void *arg) { struct intel_engine_cs *engine = arg; @@ -654,41 +705,21 @@ static int intel_breadcrumbs_signaler(void *arg) * a new client. */ rcu_read_lock(); - request = rcu_dereference(b->first_signal); - if (request) - request = i915_gem_request_get_rcu(request); + request = get_first_signal_rcu(b); rcu_read_unlock(); if (signal_complete(request)) { - local_bh_disable(); - dma_fence_signal(&request->fence); - local_bh_enable(); /* kick start the tasklets */ - - spin_lock_irq(&b->rb_lock); - - /* Wake up all other completed waiters and select the - * next bottom-half for the next user interrupt. - */ - __intel_engine_remove_wait(engine, - &request->signaling.wait); - - /* Find the next oldest signal. Note that as we have - * not been holding the lock, another client may - * have installed an even older signal than the one - * we just completed - so double check we are still - * the oldest before picking the next one. - */ - if (request == rcu_access_pointer(b->first_signal)) { - struct rb_node *rb = - rb_next(&request->signaling.node); - rcu_assign_pointer(b->first_signal, - rb ? to_signaler(rb) : NULL); + if (!test_bit(DMA_FENCE_FLAG_SIGNALED_BIT, + &request->fence.flags)) { + local_bh_disable(); + dma_fence_signal(&request->fence); + local_bh_enable(); /* kick start the tasklets */ } - rb_erase(&request->signaling.node, &b->signals); - RB_CLEAR_NODE(&request->signaling.node); - - spin_unlock_irq(&b->rb_lock); - i915_gem_request_put(request); + if (READ_ONCE(request->signaling.wait.seqno)) { + spin_lock_irq(&b->rb_lock); + __intel_engine_remove_signal(engine, request); + spin_unlock_irq(&b->rb_lock); + } /* If the engine is saturated we may be continually * processing completed requests. This angers the @@ -699,19 +730,17 @@ static int intel_breadcrumbs_signaler(void *arg) */ do_schedule = need_resched(); } + i915_gem_request_put(request); if (unlikely(do_schedule)) { if (kthread_should_park()) kthread_parkme(); - if (unlikely(kthread_should_stop())) { - i915_gem_request_put(request); + if (unlikely(kthread_should_stop())) break; - } schedule(); } - i915_gem_request_put(request); } while (1); __set_current_state(TASK_RUNNING); @@ -740,12 +769,12 @@ void intel_engine_enable_signaling(struct drm_i915_gem_request *request, if (!seqno) return; + spin_lock(&b->rb_lock); + + GEM_BUG_ON(request->signaling.wait.seqno); request->signaling.wait.tsk = b->signaler; request->signaling.wait.request = request; request->signaling.wait.seqno = seqno; - i915_gem_request_get(request); - - spin_lock(&b->rb_lock); /* First add ourselves into the list of waiters, but register our * bottom-half as the signaller thread. As per usual, only the oldest @@ -784,7 +813,7 @@ void intel_engine_enable_signaling(struct drm_i915_gem_request *request, rcu_assign_pointer(b->first_signal, request); } else { __intel_engine_remove_wait(engine, &request->signaling.wait); - i915_gem_request_put(request); + request->signaling.wait.seqno = 0; wakeup = false; } @@ -796,32 +825,17 @@ void intel_engine_enable_signaling(struct drm_i915_gem_request *request, void intel_engine_cancel_signaling(struct drm_i915_gem_request *request) { - struct intel_engine_cs *engine = request->engine; - struct intel_breadcrumbs *b = &engine->breadcrumbs; - GEM_BUG_ON(!irqs_disabled()); lockdep_assert_held(&request->lock); - GEM_BUG_ON(!request->signaling.wait.seqno); - spin_lock(&b->rb_lock); + if (READ_ONCE(request->signaling.wait.seqno)) { + struct intel_engine_cs *engine = request->engine; + struct intel_breadcrumbs *b = &engine->breadcrumbs; - if (!RB_EMPTY_NODE(&request->signaling.node)) { - if (request == rcu_access_pointer(b->first_signal)) { - struct rb_node *rb = - rb_next(&request->signaling.node); - rcu_assign_pointer(b->first_signal, - rb ? to_signaler(rb) : NULL); - } - rb_erase(&request->signaling.node, &b->signals); - RB_CLEAR_NODE(&request->signaling.node); - i915_gem_request_put(request); + spin_lock(&b->rb_lock); + __intel_engine_remove_signal(engine, request); + spin_unlock(&b->rb_lock); } - - __intel_engine_remove_wait(engine, &request->signaling.wait); - - spin_unlock(&b->rb_lock); - - request->signaling.wait.seqno = 0; } int intel_engine_init_breadcrumbs(struct intel_engine_cs *engine) diff --git a/drivers/gpu/drm/i915/intel_cdclk.c b/drivers/gpu/drm/i915/intel_cdclk.c index 5dc118f26b51..ee788d5be5e3 100644 --- a/drivers/gpu/drm/i915/intel_cdclk.c +++ b/drivers/gpu/drm/i915/intel_cdclk.c @@ -858,7 +858,7 @@ static void skl_get_cdclk(struct drm_i915_private *dev_priv, skl_dpll0_update(dev_priv, cdclk_state); - cdclk_state->cdclk = cdclk_state->ref; + cdclk_state->cdclk = cdclk_state->bypass = cdclk_state->ref; if (cdclk_state->vco == 0) goto out; @@ -1006,7 +1006,7 @@ static void skl_set_cdclk(struct drm_i915_private *dev_priv, /* Choose frequency for this cdclk */ switch (cdclk) { default: - WARN_ON(cdclk != dev_priv->cdclk.hw.ref); + WARN_ON(cdclk != dev_priv->cdclk.hw.bypass); WARN_ON(vco != 0); /* fall through */ case 308571: @@ -1085,7 +1085,7 @@ static void skl_sanitize_cdclk(struct drm_i915_private *dev_priv) /* Is PLL enabled and locked ? */ if (dev_priv->cdclk.hw.vco == 0 || - dev_priv->cdclk.hw.cdclk == dev_priv->cdclk.hw.ref) + dev_priv->cdclk.hw.cdclk == dev_priv->cdclk.hw.bypass) goto sanitize; /* DPLL okay; verify the cdclock @@ -1159,7 +1159,7 @@ void skl_uninit_cdclk(struct drm_i915_private *dev_priv) { struct intel_cdclk_state cdclk_state = dev_priv->cdclk.hw; - cdclk_state.cdclk = cdclk_state.ref; + cdclk_state.cdclk = cdclk_state.bypass; cdclk_state.vco = 0; cdclk_state.voltage_level = skl_calc_voltage_level(cdclk_state.cdclk); @@ -1199,7 +1199,7 @@ static int bxt_de_pll_vco(struct drm_i915_private *dev_priv, int cdclk) { int ratio; - if (cdclk == dev_priv->cdclk.hw.ref) + if (cdclk == dev_priv->cdclk.hw.bypass) return 0; switch (cdclk) { @@ -1224,7 +1224,7 @@ static int glk_de_pll_vco(struct drm_i915_private *dev_priv, int cdclk) { int ratio; - if (cdclk == dev_priv->cdclk.hw.ref) + if (cdclk == dev_priv->cdclk.hw.bypass) return 0; switch (cdclk) { @@ -1268,7 +1268,7 @@ static void bxt_get_cdclk(struct drm_i915_private *dev_priv, bxt_de_pll_update(dev_priv, cdclk_state); - cdclk_state->cdclk = cdclk_state->ref; + cdclk_state->cdclk = cdclk_state->bypass = cdclk_state->ref; if (cdclk_state->vco == 0) goto out; @@ -1352,7 +1352,7 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv, /* cdclk = vco / 2 / div{1,1.5,2,4} */ switch (DIV_ROUND_CLOSEST(vco, cdclk)) { default: - WARN_ON(cdclk != dev_priv->cdclk.hw.ref); + WARN_ON(cdclk != dev_priv->cdclk.hw.bypass); WARN_ON(vco != 0); /* fall through */ case 2: @@ -1378,7 +1378,7 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv, mutex_lock(&dev_priv->pcu_lock); ret = sandybridge_pcode_write_timeout(dev_priv, HSW_PCODE_DE_WRITE_FREQ_REQ, - 0x80000000, 2000); + 0x80000000, 150, 2); mutex_unlock(&dev_priv->pcu_lock); if (ret) { @@ -1417,7 +1417,7 @@ static void bxt_set_cdclk(struct drm_i915_private *dev_priv, */ ret = sandybridge_pcode_write_timeout(dev_priv, HSW_PCODE_DE_WRITE_FREQ_REQ, - cdclk_state->voltage_level, 2000); + cdclk_state->voltage_level, 150, 2); mutex_unlock(&dev_priv->pcu_lock); if (ret) { @@ -1437,7 +1437,7 @@ static void bxt_sanitize_cdclk(struct drm_i915_private *dev_priv) intel_dump_cdclk_state(&dev_priv->cdclk.hw, "Current CDCLK"); if (dev_priv->cdclk.hw.vco == 0 || - dev_priv->cdclk.hw.cdclk == dev_priv->cdclk.hw.ref) + dev_priv->cdclk.hw.cdclk == dev_priv->cdclk.hw.bypass) goto sanitize; /* DPLL okay; verify the cdclock @@ -1526,7 +1526,7 @@ void bxt_uninit_cdclk(struct drm_i915_private *dev_priv) { struct intel_cdclk_state cdclk_state = dev_priv->cdclk.hw; - cdclk_state.cdclk = cdclk_state.ref; + cdclk_state.cdclk = cdclk_state.bypass; cdclk_state.vco = 0; cdclk_state.voltage_level = bxt_calc_voltage_level(cdclk_state.cdclk); @@ -1586,7 +1586,7 @@ static void cnl_get_cdclk(struct drm_i915_private *dev_priv, cnl_cdclk_pll_update(dev_priv, cdclk_state); - cdclk_state->cdclk = cdclk_state->ref; + cdclk_state->cdclk = cdclk_state->bypass = cdclk_state->ref; if (cdclk_state->vco == 0) goto out; @@ -1672,7 +1672,7 @@ static void cnl_set_cdclk(struct drm_i915_private *dev_priv, /* cdclk = vco / 2 / div{1,2} */ switch (DIV_ROUND_CLOSEST(vco, cdclk)) { default: - WARN_ON(cdclk != dev_priv->cdclk.hw.ref); + WARN_ON(cdclk != dev_priv->cdclk.hw.bypass); WARN_ON(vco != 0); /* fall through */ case 2: @@ -1717,7 +1717,7 @@ static int cnl_cdclk_pll_vco(struct drm_i915_private *dev_priv, int cdclk) { int ratio; - if (cdclk == dev_priv->cdclk.hw.ref) + if (cdclk == dev_priv->cdclk.hw.bypass) return 0; switch (cdclk) { @@ -1744,7 +1744,7 @@ static void cnl_sanitize_cdclk(struct drm_i915_private *dev_priv) intel_dump_cdclk_state(&dev_priv->cdclk.hw, "Current CDCLK"); if (dev_priv->cdclk.hw.vco == 0 || - dev_priv->cdclk.hw.cdclk == dev_priv->cdclk.hw.ref) + dev_priv->cdclk.hw.cdclk == dev_priv->cdclk.hw.bypass) goto sanitize; /* DPLL okay; verify the cdclock @@ -1817,7 +1817,7 @@ void cnl_uninit_cdclk(struct drm_i915_private *dev_priv) { struct intel_cdclk_state cdclk_state = dev_priv->cdclk.hw; - cdclk_state.cdclk = cdclk_state.ref; + cdclk_state.cdclk = cdclk_state.bypass; cdclk_state.vco = 0; cdclk_state.voltage_level = cnl_calc_voltage_level(cdclk_state.cdclk); @@ -1858,9 +1858,10 @@ bool intel_cdclk_changed(const struct intel_cdclk_state *a, void intel_dump_cdclk_state(const struct intel_cdclk_state *cdclk_state, const char *context) { - DRM_DEBUG_DRIVER("%s %d kHz, VCO %d kHz, ref %d kHz, voltage level %d\n", + DRM_DEBUG_DRIVER("%s %d kHz, VCO %d kHz, ref %d kHz, bypass %d kHz, voltage level %d\n", context, cdclk_state->cdclk, cdclk_state->vco, - cdclk_state->ref, cdclk_state->voltage_level); + cdclk_state->ref, cdclk_state->bypass, + cdclk_state->voltage_level); } /** @@ -1952,6 +1953,14 @@ int intel_crtc_compute_min_cdclk(const struct intel_crtc_state *crtc_state) if (crtc_state->has_audio && INTEL_GEN(dev_priv) >= 9) min_cdclk = max(2 * 96000, min_cdclk); + /* + * On Valleyview some DSI panels lose (v|h)sync when the clock is lower + * than 320000KHz. + */ + if (intel_crtc_has_type(crtc_state, INTEL_OUTPUT_DSI) && + IS_VALLEYVIEW(dev_priv)) + min_cdclk = max(320000, min_cdclk); + if (min_cdclk > dev_priv->max_cdclk_freq) { DRM_DEBUG_KMS("required cdclk (%d kHz) exceeds max (%d kHz)\n", min_cdclk, dev_priv->max_cdclk_freq); @@ -2346,6 +2355,30 @@ static int cnp_rawclk(struct drm_i915_private *dev_priv) return divider + fraction; } +static int icp_rawclk(struct drm_i915_private *dev_priv) +{ + u32 rawclk; + int divider, numerator, denominator, frequency; + + if (I915_READ(SFUSE_STRAP) & SFUSE_STRAP_RAW_FREQUENCY) { + frequency = 24000; + divider = 23; + numerator = 0; + denominator = 0; + } else { + frequency = 19200; + divider = 18; + numerator = 1; + denominator = 4; + } + + rawclk = CNP_RAWCLK_DIV(divider) | ICP_RAWCLK_NUM(numerator) | + ICP_RAWCLK_DEN(denominator); + + I915_WRITE(PCH_RAWCLK_FREQ, rawclk); + return frequency; +} + static int pch_rawclk(struct drm_i915_private *dev_priv) { return (I915_READ(PCH_RAWCLK_FREQ) & RAWCLK_FREQ_MASK) * 1000; @@ -2393,8 +2426,9 @@ static int g4x_hrawclk(struct drm_i915_private *dev_priv) */ void intel_update_rawclk(struct drm_i915_private *dev_priv) { - - if (HAS_PCH_CNP(dev_priv)) + if (HAS_PCH_ICP(dev_priv)) + dev_priv->rawclk_freq = icp_rawclk(dev_priv); + else if (HAS_PCH_CNP(dev_priv)) dev_priv->rawclk_freq = cnp_rawclk(dev_priv); else if (HAS_PCH_SPLIT(dev_priv)) dev_priv->rawclk_freq = pch_rawclk(dev_priv); diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c index 7fe4aac0facc..41e6c75a7f3c 100644 --- a/drivers/gpu/drm/i915/intel_csr.c +++ b/drivers/gpu/drm/i915/intel_csr.c @@ -37,8 +37,9 @@ #define I915_CSR_GLK "i915/glk_dmc_ver1_04.bin" #define GLK_CSR_VERSION_REQUIRED CSR_VERSION(1, 4) -#define I915_CSR_CNL "i915/cnl_dmc_ver1_06.bin" -#define CNL_CSR_VERSION_REQUIRED CSR_VERSION(1, 6) +#define I915_CSR_CNL "i915/cnl_dmc_ver1_07.bin" +MODULE_FIRMWARE(I915_CSR_CNL); +#define CNL_CSR_VERSION_REQUIRED CSR_VERSION(1, 7) #define I915_CSR_KBL "i915/kbl_dmc_ver1_04.bin" MODULE_FIRMWARE(I915_CSR_KBL); diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c index f51645a08dca..d9b52fe82932 100644 --- a/drivers/gpu/drm/i915/intel_ddi.c +++ b/drivers/gpu/drm/i915/intel_ddi.c @@ -2404,6 +2404,48 @@ static void intel_enable_ddi_hdmi(struct intel_encoder *encoder, crtc_state->hdmi_high_tmds_clock_ratio, crtc_state->hdmi_scrambling); + /* Display WA #1143: skl,kbl,cfl */ + if (IS_GEN9_BC(dev_priv)) { + /* + * For some reason these chicken bits have been + * stuffed into a transcoder register, event though + * the bits affect a specific DDI port rather than + * a specific transcoder. + */ + static const enum transcoder port_to_transcoder[] = { + [PORT_A] = TRANSCODER_EDP, + [PORT_B] = TRANSCODER_A, + [PORT_C] = TRANSCODER_B, + [PORT_D] = TRANSCODER_C, + [PORT_E] = TRANSCODER_A, + }; + enum transcoder transcoder = port_to_transcoder[port]; + u32 val; + + val = I915_READ(CHICKEN_TRANS(transcoder)); + + if (port == PORT_E) + val |= DDIE_TRAINING_OVERRIDE_ENABLE | + DDIE_TRAINING_OVERRIDE_VALUE; + else + val |= DDI_TRAINING_OVERRIDE_ENABLE | + DDI_TRAINING_OVERRIDE_VALUE; + + I915_WRITE(CHICKEN_TRANS(transcoder), val); + POSTING_READ(CHICKEN_TRANS(transcoder)); + + udelay(1); + + if (port == PORT_E) + val &= ~(DDIE_TRAINING_OVERRIDE_ENABLE | + DDIE_TRAINING_OVERRIDE_VALUE); + else + val &= ~(DDI_TRAINING_OVERRIDE_ENABLE | + DDI_TRAINING_OVERRIDE_VALUE); + + I915_WRITE(CHICKEN_TRANS(transcoder), val); + } + /* In HDMI/DVI mode, the port width, and swing/emphasis values * are ignored so nothing special needs to be done besides * enabling the port. @@ -2868,6 +2910,10 @@ void intel_ddi_init(struct drm_i915_private *dev_priv, enum port port) intel_dig_port->ddi_io_power_domain = POWER_DOMAIN_PORT_DDI_E_IO; break; + case PORT_F: + intel_dig_port->ddi_io_power_domain = + POWER_DOMAIN_PORT_DDI_F_IO; + break; default: MISSING_CASE(port); } diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c index d28592e43512..a2c16140169f 100644 --- a/drivers/gpu/drm/i915/intel_device_info.c +++ b/drivers/gpu/drm/i915/intel_device_info.c @@ -56,6 +56,7 @@ static const char * const platform_names[] = { PLATFORM_NAME(GEMINILAKE), PLATFORM_NAME(COFFEELAKE), PLATFORM_NAME(CANNONLAKE), + PLATFORM_NAME(ICELAKE), }; #undef PLATFORM_NAME diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h index 49cb27bd04c1..9542018d11d0 100644 --- a/drivers/gpu/drm/i915/intel_device_info.h +++ b/drivers/gpu/drm/i915/intel_device_info.h @@ -69,6 +69,8 @@ enum intel_platform { INTEL_COFFEELAKE, /* gen10 */ INTEL_CANNONLAKE, + /* gen11 */ + INTEL_ICELAKE, INTEL_MAX_PLATFORMS }; diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index f288bcc7be22..e9bba2ab4904 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -2387,6 +2387,20 @@ static unsigned int intel_fb_modifier_to_tiling(uint64_t fb_modifier) } } +/* + * From the Sky Lake PRM: + * "The Color Control Surface (CCS) contains the compression status of + * the cache-line pairs. The compression state of the cache-line pair + * is specified by 2 bits in the CCS. Each CCS cache-line represents + * an area on the main surface of 16 x16 sets of 128 byte Y-tiled + * cache-line-pairs. CCS is always Y tiled." + * + * Since cache line pairs refers to horizontally adjacent cache lines, + * each cache line in the CCS corresponds to an area of 32x16 cache + * lines on the main surface. Since each pixel is 4 bytes, this gives + * us a ratio of one byte in the CCS for each 8x16 pixels in the + * main surface. + */ static const struct drm_format_info ccs_formats[] = { { .format = DRM_FORMAT_XRGB8888, .depth = 24, .num_planes = 2, .cpp = { 4, 1, }, .hsub = 8, .vsub = 16, }, { .format = DRM_FORMAT_XBGR8888, .depth = 24, .num_planes = 2, .cpp = { 4, 1, }, .hsub = 8, .vsub = 16, }, @@ -2917,14 +2931,19 @@ static bool skl_check_main_ccs_coordinates(struct intel_plane_state *plane_state return true; } -static int skl_check_main_surface(struct intel_plane_state *plane_state) +static int skl_check_main_surface(const struct intel_crtc_state *crtc_state, + struct intel_plane_state *plane_state) { + struct drm_i915_private *dev_priv = + to_i915(plane_state->base.plane->dev); const struct drm_framebuffer *fb = plane_state->base.fb; unsigned int rotation = plane_state->base.rotation; int x = plane_state->base.src.x1 >> 16; int y = plane_state->base.src.y1 >> 16; int w = drm_rect_width(&plane_state->base.src) >> 16; int h = drm_rect_height(&plane_state->base.src) >> 16; + int dst_x = plane_state->base.dst.x1; + int pipe_src_w = crtc_state->pipe_src_w; int max_width = skl_max_plane_width(fb, 0, rotation); int max_height = 4096; u32 alignment, offset, aux_offset = plane_state->aux.offset; @@ -2935,6 +2954,24 @@ static int skl_check_main_surface(struct intel_plane_state *plane_state) return -EINVAL; } + /* + * Display WA #1175: cnl,glk + * Planes other than the cursor may cause FIFO underflow and display + * corruption if starting less than 4 pixels from the right edge of + * the screen. + * Besides the above WA fix the similar problem, where planes other + * than the cursor ending less than 4 pixels from the left edge of the + * screen may cause FIFO underflow and display corruption. + */ + if ((IS_GEMINILAKE(dev_priv) || IS_CANNONLAKE(dev_priv)) && + (dst_x + w < 4 || dst_x > pipe_src_w - 4)) { + DRM_DEBUG_KMS("requested plane X %s position %d invalid (valid range %d-%d)\n", + dst_x + w < 4 ? "end" : "start", + dst_x + w < 4 ? dst_x + w : dst_x, + 4, pipe_src_w - 4); + return -ERANGE; + } + intel_add_fb_offsets(&x, &y, plane_state, 0); offset = intel_compute_tile_offset(&x, &y, plane_state, 0); alignment = intel_surf_alignment(fb, 0); @@ -3027,6 +3064,7 @@ static int skl_check_nv12_aux_surface(struct intel_plane_state *plane_state) static int skl_check_ccs_aux_surface(struct intel_plane_state *plane_state) { struct intel_plane *plane = to_intel_plane(plane_state->base.plane); + struct drm_i915_private *dev_priv = to_i915(plane->base.dev); struct intel_crtc *crtc = to_intel_crtc(plane_state->base.crtc); const struct drm_framebuffer *fb = plane_state->base.fb; int src_x = plane_state->base.src.x1 >> 16; @@ -3037,17 +3075,8 @@ static int skl_check_ccs_aux_surface(struct intel_plane_state *plane_state) int y = src_y / vsub; u32 offset; - switch (plane->id) { - case PLANE_PRIMARY: - case PLANE_SPRITE0: - break; - default: - DRM_DEBUG_KMS("RC support only on plane 1 and 2\n"); - return -EINVAL; - } - - if (crtc->pipe == PIPE_C) { - DRM_DEBUG_KMS("No RC support on pipe C\n"); + if (!skl_plane_has_ccs(dev_priv, crtc->pipe, plane->id)) { + DRM_DEBUG_KMS("No RC support on %s\n", plane->base.name); return -EINVAL; } @@ -3067,7 +3096,8 @@ static int skl_check_ccs_aux_surface(struct intel_plane_state *plane_state) return 0; } -int skl_check_plane_surface(struct intel_plane_state *plane_state) +int skl_check_plane_surface(const struct intel_crtc_state *crtc_state, + struct intel_plane_state *plane_state) { const struct drm_framebuffer *fb = plane_state->base.fb; unsigned int rotation = plane_state->base.rotation; @@ -3107,7 +3137,7 @@ int skl_check_plane_surface(struct intel_plane_state *plane_state) plane_state->aux.y = 0; } - ret = skl_check_main_surface(plane_state); + ret = skl_check_main_surface(crtc_state, plane_state); if (ret) return ret; @@ -4757,7 +4787,7 @@ static int skl_update_scaler_plane(struct intel_crtc_state *crtc_state, return ret; /* check colorkey */ - if (plane_state->ckey.flags != I915_SET_COLORKEY_NONE) { + if (plane_state->ckey.flags) { DRM_DEBUG_KMS("[PLANE:%d:%s] scaling with color key not allowed", intel_plane->base.base.id, intel_plane->base.name); @@ -5641,6 +5671,8 @@ enum intel_display_power_domain intel_port_to_power_domain(enum port port) return POWER_DOMAIN_PORT_DDI_D_LANES; case PORT_E: return POWER_DOMAIN_PORT_DDI_E_LANES; + case PORT_F: + return POWER_DOMAIN_PORT_DDI_F_LANES; default: MISSING_CASE(port); return POWER_DOMAIN_PORT_OTHER; @@ -8495,7 +8527,10 @@ skylake_get_initial_plane_config(struct intel_crtc *crtc, val = I915_READ(PLANE_CTL(pipe, plane_id)); - pixel_format = val & PLANE_CTL_FORMAT_MASK; + if (INTEL_GEN(dev_priv) >= 11) + pixel_format = val & ICL_PLANE_CTL_FORMAT_MASK; + else + pixel_format = val & PLANE_CTL_FORMAT_MASK; if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv)) { alpha = I915_READ(PLANE_COLOR_CTL(pipe, plane_id)); @@ -12519,7 +12554,13 @@ static int do_rps_boost(struct wait_queue_entry *_wait, struct wait_rps_boost *wait = container_of(_wait, typeof(*wait), wait); struct drm_i915_gem_request *rq = wait->request; - gen6_rps_boost(rq, NULL); + /* + * If we missed the vblank, but the request is already running it + * is reasonable to assume that it will complete before the next + * vblank without our intervention, so leave RPS alone. + */ + if (!i915_gem_request_started(rq)) + gen6_rps_boost(rq, NULL); i915_gem_request_put(rq); drm_crtc_vblank_put(wait->crtc); @@ -12747,7 +12788,7 @@ intel_check_primary_plane(struct intel_plane *plane, if (INTEL_GEN(dev_priv) >= 9) { /* use scaler when colorkey is not required */ - if (state->ckey.flags == I915_SET_COLORKEY_NONE) { + if (!state->ckey.flags) { min_scale = 1; max_scale = skl_max_scale(to_intel_crtc(crtc), crtc_state); } @@ -12766,7 +12807,7 @@ intel_check_primary_plane(struct intel_plane *plane, return 0; if (INTEL_GEN(dev_priv) >= 9) { - ret = skl_check_plane_surface(state); + ret = skl_check_plane_surface(crtc_state, state); if (ret) return ret; @@ -12944,8 +12985,6 @@ static bool intel_primary_plane_format_mod_supported(struct drm_plane *plane, return i965_mod_supported(format, modifier); else return i8xx_mod_supported(format, modifier); - - unreachable(); } static bool intel_cursor_plane_format_mod_supported(struct drm_plane *plane, @@ -13153,21 +13192,14 @@ intel_primary_plane_create(struct drm_i915_private *dev_priv, enum pipe pipe) else primary->i9xx_plane = (enum i9xx_plane_id) pipe; primary->id = PLANE_PRIMARY; - primary->frontbuffer_bit = INTEL_FRONTBUFFER_PRIMARY(pipe); + primary->frontbuffer_bit = INTEL_FRONTBUFFER(pipe, primary->id); primary->check_plane = intel_check_primary_plane; - if (INTEL_GEN(dev_priv) >= 10) { + if (INTEL_GEN(dev_priv) >= 9) { intel_primary_formats = skl_primary_formats; num_formats = ARRAY_SIZE(skl_primary_formats); - modifiers = skl_format_modifiers_ccs; - primary->update_plane = skl_update_plane; - primary->disable_plane = skl_disable_plane; - primary->get_hw_state = skl_plane_get_hw_state; - } else if (INTEL_GEN(dev_priv) >= 9) { - intel_primary_formats = skl_primary_formats; - num_formats = ARRAY_SIZE(skl_primary_formats); - if (pipe < PIPE_C) + if (skl_plane_has_ccs(dev_priv, pipe, PLANE_PRIMARY)) modifiers = skl_format_modifiers_ccs; else modifiers = skl_format_modifiers_noccs; @@ -13281,7 +13313,7 @@ intel_cursor_plane_create(struct drm_i915_private *dev_priv, cursor->pipe = pipe; cursor->i9xx_plane = (enum i9xx_plane_id) pipe; cursor->id = PLANE_CURSOR; - cursor->frontbuffer_bit = INTEL_FRONTBUFFER_CURSOR(pipe); + cursor->frontbuffer_bit = INTEL_FRONTBUFFER(pipe, cursor->id); if (IS_I845G(dev_priv) || IS_I865G(dev_priv)) { cursor->update_plane = i845_update_cursor; @@ -13597,7 +13629,7 @@ static void intel_setup_outputs(struct drm_i915_private *dev_priv) if (found || IS_GEN9_BC(dev_priv)) intel_ddi_init(dev_priv, PORT_A); - /* DDI B, C and D detection is indicated by the SFUSE_STRAP + /* DDI B, C, D, and F detection is indicated by the SFUSE_STRAP * register */ found = I915_READ(SFUSE_STRAP); @@ -13607,6 +13639,8 @@ static void intel_setup_outputs(struct drm_i915_private *dev_priv) intel_ddi_init(dev_priv, PORT_C); if (found & SFUSE_STRAP_DDID_DETECTED) intel_ddi_init(dev_priv, PORT_D); + if (found & SFUSE_STRAP_DDIF_DETECTED) + intel_ddi_init(dev_priv, PORT_F); /* * On SKL we don't have a way to detect DDI-E so we rely on VBT. */ diff --git a/drivers/gpu/drm/i915/intel_display.h b/drivers/gpu/drm/i915/intel_display.h index a0d2b6169361..c4042e342f50 100644 --- a/drivers/gpu/drm/i915/intel_display.h +++ b/drivers/gpu/drm/i915/intel_display.h @@ -119,6 +119,7 @@ enum port { PORT_C, PORT_D, PORT_E, + PORT_F, I915_MAX_PORTS }; @@ -156,11 +157,13 @@ enum intel_display_power_domain { POWER_DOMAIN_PORT_DDI_C_LANES, POWER_DOMAIN_PORT_DDI_D_LANES, POWER_DOMAIN_PORT_DDI_E_LANES, + POWER_DOMAIN_PORT_DDI_F_LANES, POWER_DOMAIN_PORT_DDI_A_IO, POWER_DOMAIN_PORT_DDI_B_IO, POWER_DOMAIN_PORT_DDI_C_IO, POWER_DOMAIN_PORT_DDI_D_IO, POWER_DOMAIN_PORT_DDI_E_IO, + POWER_DOMAIN_PORT_DDI_F_IO, POWER_DOMAIN_PORT_DSI, POWER_DOMAIN_PORT_CRT, POWER_DOMAIN_PORT_OTHER, @@ -171,6 +174,7 @@ enum intel_display_power_domain { POWER_DOMAIN_AUX_B, POWER_DOMAIN_AUX_C, POWER_DOMAIN_AUX_D, + POWER_DOMAIN_AUX_F, POWER_DOMAIN_GMBUS, POWER_DOMAIN_MODESET, POWER_DOMAIN_GT_IRQ, diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index 35c5299feab6..8503d182921b 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -155,6 +155,28 @@ static void intel_dp_set_sink_rates(struct intel_dp *intel_dp) intel_dp->num_sink_rates = i; } +/* Get length of rates array potentially limited by max_rate. */ +static int intel_dp_rate_limit_len(const int *rates, int len, int max_rate) +{ + int i; + + /* Limit results by potentially reduced max rate */ + for (i = 0; i < len; i++) { + if (rates[len - i - 1] <= max_rate) + return len - i; + } + + return 0; +} + +/* Get length of common rates array potentially limited by max_rate. */ +static int intel_dp_common_len_rate_limit(const struct intel_dp *intel_dp, + int max_rate) +{ + return intel_dp_rate_limit_len(intel_dp->common_rates, + intel_dp->num_common_rates, max_rate); +} + /* Theoretical max between source and sink */ static int intel_dp_max_common_rate(struct intel_dp *intel_dp) { @@ -218,15 +240,38 @@ intel_dp_downstream_max_dotclock(struct intel_dp *intel_dp) return max_dotclk; } +static int cnl_max_source_rate(struct intel_dp *intel_dp) +{ + struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); + struct drm_i915_private *dev_priv = to_i915(dig_port->base.base.dev); + enum port port = dig_port->base.port; + + u32 voltage = I915_READ(CNL_PORT_COMP_DW3) & VOLTAGE_INFO_MASK; + + /* Low voltage SKUs are limited to max of 5.4G */ + if (voltage == VOLTAGE_INFO_0_85V) + return 540000; + + /* For this SKU 8.1G is supported in all ports */ + if (IS_CNL_WITH_PORT_F(dev_priv)) + return 810000; + + /* For other SKUs, max rate on ports A and B is 5.4G */ + if (port == PORT_A || port == PORT_D) + return 540000; + + return 810000; +} + static void intel_dp_set_source_rates(struct intel_dp *intel_dp) { struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); struct drm_i915_private *dev_priv = to_i915(dig_port->base.base.dev); - enum port port = dig_port->base.port; + const struct ddi_vbt_port_info *info = + &dev_priv->vbt.ddi_port_info[dig_port->base.port]; const int *source_rates; - int size; - u32 voltage; + int size, max_rate = 0, vbt_max_rate = info->dp_max_link_rate; /* This should only be done once */ WARN_ON(intel_dp->source_rates || intel_dp->num_source_rates); @@ -237,10 +282,7 @@ intel_dp_set_source_rates(struct intel_dp *intel_dp) } else if (IS_CANNONLAKE(dev_priv)) { source_rates = cnl_rates; size = ARRAY_SIZE(cnl_rates); - voltage = I915_READ(CNL_PORT_COMP_DW3) & VOLTAGE_INFO_MASK; - if (port == PORT_A || port == PORT_D || - voltage == VOLTAGE_INFO_0_85V) - size -= 2; + max_rate = cnl_max_source_rate(intel_dp); } else if (IS_GEN9_BC(dev_priv)) { source_rates = skl_rates; size = ARRAY_SIZE(skl_rates); @@ -253,6 +295,14 @@ intel_dp_set_source_rates(struct intel_dp *intel_dp) size = ARRAY_SIZE(default_rates) - 1; } + if (max_rate && vbt_max_rate) + max_rate = min(max_rate, vbt_max_rate); + else if (vbt_max_rate) + max_rate = vbt_max_rate; + + if (max_rate) + size = intel_dp_rate_limit_len(source_rates, size, max_rate); + intel_dp->source_rates = source_rates; intel_dp->num_source_rates = size; } @@ -309,22 +359,6 @@ static void intel_dp_set_common_rates(struct intel_dp *intel_dp) } } -/* get length of common rates potentially limited by max_rate */ -static int intel_dp_common_len_rate_limit(struct intel_dp *intel_dp, - int max_rate) -{ - const int *common_rates = intel_dp->common_rates; - int i, common_len = intel_dp->num_common_rates; - - /* Limit results by potentially reduced max rate */ - for (i = 0; i < common_len; i++) { - if (common_rates[common_len - i - 1] <= max_rate) - return common_len - i; - } - - return 0; -} - static bool intel_dp_link_params_valid(struct intel_dp *intel_dp, int link_rate, uint8_t lane_count) { @@ -794,7 +828,8 @@ static void intel_pps_get_registers(struct intel_dp *intel_dp, regs->pp_stat = PP_STATUS(pps_idx); regs->pp_on = PP_ON_DELAYS(pps_idx); regs->pp_off = PP_OFF_DELAYS(pps_idx); - if (!IS_GEN9_LP(dev_priv) && !HAS_PCH_CNP(dev_priv)) + if (!IS_GEN9_LP(dev_priv) && !HAS_PCH_CNP(dev_priv) && + !HAS_PCH_ICP(dev_priv)) regs->pp_div = PP_DIVISOR(pps_idx); } @@ -1298,6 +1333,9 @@ static enum port intel_aux_port(struct drm_i915_private *dev_priv, case DP_AUX_D: aux_port = PORT_D; break; + case DP_AUX_F: + aux_port = PORT_F; + break; default: MISSING_CASE(info->alternate_aux_channel); aux_port = PORT_A; @@ -1378,6 +1416,7 @@ static i915_reg_t skl_aux_ctl_reg(struct drm_i915_private *dev_priv, case PORT_B: case PORT_C: case PORT_D: + case PORT_F: return DP_AUX_CH_CTL(port); default: MISSING_CASE(port); @@ -1393,6 +1432,7 @@ static i915_reg_t skl_aux_data_reg(struct drm_i915_private *dev_priv, case PORT_B: case PORT_C: case PORT_D: + case PORT_F: return DP_AUX_CH_DATA(port, index); default: MISSING_CASE(port); @@ -4455,173 +4495,174 @@ edp_detect(struct intel_dp *intel_dp) return status; } -static bool ibx_digital_port_connected(struct drm_i915_private *dev_priv, - struct intel_digital_port *port) +static bool ibx_digital_port_connected(struct intel_encoder *encoder) { + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); u32 bit; - switch (port->base.port) { - case PORT_B: + switch (encoder->hpd_pin) { + case HPD_PORT_B: bit = SDE_PORTB_HOTPLUG; break; - case PORT_C: + case HPD_PORT_C: bit = SDE_PORTC_HOTPLUG; break; - case PORT_D: + case HPD_PORT_D: bit = SDE_PORTD_HOTPLUG; break; default: - MISSING_CASE(port->base.port); + MISSING_CASE(encoder->hpd_pin); return false; } return I915_READ(SDEISR) & bit; } -static bool cpt_digital_port_connected(struct drm_i915_private *dev_priv, - struct intel_digital_port *port) +static bool cpt_digital_port_connected(struct intel_encoder *encoder) { + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); u32 bit; - switch (port->base.port) { - case PORT_B: + switch (encoder->hpd_pin) { + case HPD_PORT_B: bit = SDE_PORTB_HOTPLUG_CPT; break; - case PORT_C: + case HPD_PORT_C: bit = SDE_PORTC_HOTPLUG_CPT; break; - case PORT_D: + case HPD_PORT_D: bit = SDE_PORTD_HOTPLUG_CPT; break; default: - MISSING_CASE(port->base.port); + MISSING_CASE(encoder->hpd_pin); return false; } return I915_READ(SDEISR) & bit; } -static bool spt_digital_port_connected(struct drm_i915_private *dev_priv, - struct intel_digital_port *port) +static bool spt_digital_port_connected(struct intel_encoder *encoder) { + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); u32 bit; - switch (port->base.port) { - case PORT_A: + switch (encoder->hpd_pin) { + case HPD_PORT_A: bit = SDE_PORTA_HOTPLUG_SPT; break; - case PORT_E: + case HPD_PORT_E: bit = SDE_PORTE_HOTPLUG_SPT; break; default: - return cpt_digital_port_connected(dev_priv, port); + return cpt_digital_port_connected(encoder); } return I915_READ(SDEISR) & bit; } -static bool g4x_digital_port_connected(struct drm_i915_private *dev_priv, - struct intel_digital_port *port) +static bool g4x_digital_port_connected(struct intel_encoder *encoder) { + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); u32 bit; - switch (port->base.port) { - case PORT_B: + switch (encoder->hpd_pin) { + case HPD_PORT_B: bit = PORTB_HOTPLUG_LIVE_STATUS_G4X; break; - case PORT_C: + case HPD_PORT_C: bit = PORTC_HOTPLUG_LIVE_STATUS_G4X; break; - case PORT_D: + case HPD_PORT_D: bit = PORTD_HOTPLUG_LIVE_STATUS_G4X; break; default: - MISSING_CASE(port->base.port); + MISSING_CASE(encoder->hpd_pin); return false; } return I915_READ(PORT_HOTPLUG_STAT) & bit; } -static bool gm45_digital_port_connected(struct drm_i915_private *dev_priv, - struct intel_digital_port *port) +static bool gm45_digital_port_connected(struct intel_encoder *encoder) { + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); u32 bit; - switch (port->base.port) { - case PORT_B: + switch (encoder->hpd_pin) { + case HPD_PORT_B: bit = PORTB_HOTPLUG_LIVE_STATUS_GM45; break; - case PORT_C: + case HPD_PORT_C: bit = PORTC_HOTPLUG_LIVE_STATUS_GM45; break; - case PORT_D: + case HPD_PORT_D: bit = PORTD_HOTPLUG_LIVE_STATUS_GM45; break; default: - MISSING_CASE(port->base.port); + MISSING_CASE(encoder->hpd_pin); return false; } return I915_READ(PORT_HOTPLUG_STAT) & bit; } -static bool ilk_digital_port_connected(struct drm_i915_private *dev_priv, - struct intel_digital_port *port) +static bool ilk_digital_port_connected(struct intel_encoder *encoder) { - if (port->base.port == PORT_A) + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + + if (encoder->hpd_pin == HPD_PORT_A) return I915_READ(DEISR) & DE_DP_A_HOTPLUG; else - return ibx_digital_port_connected(dev_priv, port); + return ibx_digital_port_connected(encoder); } -static bool snb_digital_port_connected(struct drm_i915_private *dev_priv, - struct intel_digital_port *port) +static bool snb_digital_port_connected(struct intel_encoder *encoder) { - if (port->base.port == PORT_A) + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + + if (encoder->hpd_pin == HPD_PORT_A) return I915_READ(DEISR) & DE_DP_A_HOTPLUG; else - return cpt_digital_port_connected(dev_priv, port); + return cpt_digital_port_connected(encoder); } -static bool ivb_digital_port_connected(struct drm_i915_private *dev_priv, - struct intel_digital_port *port) +static bool ivb_digital_port_connected(struct intel_encoder *encoder) { - if (port->base.port == PORT_A) + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + + if (encoder->hpd_pin == HPD_PORT_A) return I915_READ(DEISR) & DE_DP_A_HOTPLUG_IVB; else - return cpt_digital_port_connected(dev_priv, port); + return cpt_digital_port_connected(encoder); } -static bool bdw_digital_port_connected(struct drm_i915_private *dev_priv, - struct intel_digital_port *port) +static bool bdw_digital_port_connected(struct intel_encoder *encoder) { - if (port->base.port == PORT_A) + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + + if (encoder->hpd_pin == HPD_PORT_A) return I915_READ(GEN8_DE_PORT_ISR) & GEN8_PORT_DP_A_HOTPLUG; else - return cpt_digital_port_connected(dev_priv, port); + return cpt_digital_port_connected(encoder); } -static bool bxt_digital_port_connected(struct drm_i915_private *dev_priv, - struct intel_digital_port *intel_dig_port) +static bool bxt_digital_port_connected(struct intel_encoder *encoder) { - struct intel_encoder *intel_encoder = &intel_dig_port->base; - enum port port; + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); u32 bit; - port = intel_hpd_pin_to_port(intel_encoder->hpd_pin); - switch (port) { - case PORT_A: + switch (encoder->hpd_pin) { + case HPD_PORT_A: bit = BXT_DE_PORT_HP_DDIA; break; - case PORT_B: + case HPD_PORT_B: bit = BXT_DE_PORT_HP_DDIB; break; - case PORT_C: + case HPD_PORT_C: bit = BXT_DE_PORT_HP_DDIC; break; default: - MISSING_CASE(port); + MISSING_CASE(encoder->hpd_pin); return false; } @@ -4630,33 +4671,33 @@ static bool bxt_digital_port_connected(struct drm_i915_private *dev_priv, /* * intel_digital_port_connected - is the specified port connected? - * @dev_priv: i915 private structure - * @port: the port to test + * @encoder: intel_encoder * - * Return %true if @port is connected, %false otherwise. + * Return %true if port is connected, %false otherwise. */ -bool intel_digital_port_connected(struct drm_i915_private *dev_priv, - struct intel_digital_port *port) +bool intel_digital_port_connected(struct intel_encoder *encoder) { + struct drm_i915_private *dev_priv = to_i915(encoder->base.dev); + if (HAS_GMCH_DISPLAY(dev_priv)) { if (IS_GM45(dev_priv)) - return gm45_digital_port_connected(dev_priv, port); + return gm45_digital_port_connected(encoder); else - return g4x_digital_port_connected(dev_priv, port); + return g4x_digital_port_connected(encoder); } if (IS_GEN5(dev_priv)) - return ilk_digital_port_connected(dev_priv, port); + return ilk_digital_port_connected(encoder); else if (IS_GEN6(dev_priv)) - return snb_digital_port_connected(dev_priv, port); + return snb_digital_port_connected(encoder); else if (IS_GEN7(dev_priv)) - return ivb_digital_port_connected(dev_priv, port); + return ivb_digital_port_connected(encoder); else if (IS_GEN8(dev_priv)) - return bdw_digital_port_connected(dev_priv, port); + return bdw_digital_port_connected(encoder); else if (IS_GEN9_LP(dev_priv)) - return bxt_digital_port_connected(dev_priv, port); + return bxt_digital_port_connected(encoder); else - return spt_digital_port_connected(dev_priv, port); + return spt_digital_port_connected(encoder); } static struct edid * @@ -4715,8 +4756,7 @@ intel_dp_long_pulse(struct intel_connector *connector) /* Can't disconnect eDP, but you can close the lid... */ if (intel_dp_is_edp(intel_dp)) status = edp_detect(intel_dp); - else if (intel_digital_port_connected(dev_priv, - dp_to_dig_port(intel_dp))) + else if (intel_digital_port_connected(&dp_to_dig_port(intel_dp)->base)) status = intel_dp_detect_dpcd(intel_dp); else status = connector_status_disconnected; @@ -5227,7 +5267,8 @@ intel_pps_readout_hw_state(struct intel_dp *intel_dp, struct edp_power_seq *seq) pp_on = I915_READ(regs.pp_on); pp_off = I915_READ(regs.pp_off); - if (!IS_GEN9_LP(dev_priv) && !HAS_PCH_CNP(dev_priv)) { + if (!IS_GEN9_LP(dev_priv) && !HAS_PCH_CNP(dev_priv) && + !HAS_PCH_ICP(dev_priv)) { I915_WRITE(regs.pp_ctrl, pp_ctl); pp_div = I915_READ(regs.pp_div); } @@ -5245,7 +5286,8 @@ intel_pps_readout_hw_state(struct intel_dp *intel_dp, struct edp_power_seq *seq) seq->t10 = (pp_off & PANEL_POWER_DOWN_DELAY_MASK) >> PANEL_POWER_DOWN_DELAY_SHIFT; - if (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv)) { + if (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv) || + HAS_PCH_ICP(dev_priv)) { seq->t11_t12 = ((pp_ctl & BXT_POWER_CYCLE_DELAY_MASK) >> BXT_POWER_CYCLE_DELAY_SHIFT) * 1000; } else { @@ -5416,7 +5458,8 @@ intel_dp_init_panel_power_sequencer_registers(struct intel_dp *intel_dp, (seq->t10 << PANEL_POWER_DOWN_DELAY_SHIFT); /* Compute the divisor for the pp clock, simply match the Bspec * formula. */ - if (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv)) { + if (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv) || + HAS_PCH_ICP(dev_priv)) { pp_div = I915_READ(regs.pp_ctrl); pp_div &= ~BXT_POWER_CYCLE_DELAY_MASK; pp_div |= (DIV_ROUND_UP(seq->t11_t12, 1000) @@ -5442,7 +5485,8 @@ intel_dp_init_panel_power_sequencer_registers(struct intel_dp *intel_dp, I915_WRITE(regs.pp_on, pp_on); I915_WRITE(regs.pp_off, pp_off); - if (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv)) + if (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv) || + HAS_PCH_ICP(dev_priv)) I915_WRITE(regs.pp_ctrl, pp_div); else I915_WRITE(regs.pp_div, pp_div); @@ -5450,7 +5494,8 @@ intel_dp_init_panel_power_sequencer_registers(struct intel_dp *intel_dp, DRM_DEBUG_KMS("panel power sequencer register settings: PP_ON %#x, PP_OFF %#x, PP_DIV %#x\n", I915_READ(regs.pp_on), I915_READ(regs.pp_off), - (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv)) ? + (IS_GEN9_LP(dev_priv) || HAS_PCH_CNP(dev_priv) || + HAS_PCH_ICP(dev_priv)) ? (I915_READ(regs.pp_ctrl) & BXT_POWER_CYCLE_DELAY_MASK) : I915_READ(regs.pp_div)); } @@ -5970,8 +6015,10 @@ intel_dp_init_connector_port_info(struct intel_digital_port *intel_dig_port) { struct intel_encoder *encoder = &intel_dig_port->base; struct intel_dp *intel_dp = &intel_dig_port->dp; + struct intel_encoder *intel_encoder = &intel_dig_port->base; + struct drm_i915_private *dev_priv = to_i915(intel_encoder->base.dev); - encoder->hpd_pin = intel_hpd_pin(encoder->port); + encoder->hpd_pin = intel_hpd_pin_default(dev_priv, encoder->port); switch (encoder->port) { case PORT_A: @@ -5990,6 +6037,9 @@ intel_dp_init_connector_port_info(struct intel_digital_port *intel_dig_port) /* FIXME: Check VBT for actual wiring of PORT E */ intel_dp->aux_power_domain = POWER_DOMAIN_AUX_D; break; + case PORT_F: + intel_dp->aux_power_domain = POWER_DOMAIN_AUX_F; + break; default: MISSING_CASE(encoder->port); } @@ -6116,7 +6166,8 @@ intel_dp_init_connector(struct intel_digital_port *intel_dig_port, /* init MST on ports that can support it */ if (HAS_DP_MST(dev_priv) && !intel_dp_is_edp(intel_dp) && - (port == PORT_B || port == PORT_C || port == PORT_D)) + (port == PORT_B || port == PORT_C || + port == PORT_D || port == PORT_F)) intel_dp_mst_encoder_init(intel_dig_port, intel_connector->base.base.id); diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index 30f791f89d64..71355fdec94a 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -1507,7 +1507,8 @@ u32 skl_plane_ctl(const struct intel_crtc_state *crtc_state, const struct intel_plane_state *plane_state); u32 skl_plane_stride(const struct drm_framebuffer *fb, int plane, unsigned int rotation); -int skl_check_plane_surface(struct intel_plane_state *plane_state); +int skl_check_plane_surface(const struct intel_crtc_state *crtc_state, + struct intel_plane_state *plane_state); int i9xx_check_plane_surface(struct intel_plane_state *plane_state); /* intel_csr.c */ @@ -1590,8 +1591,7 @@ static inline unsigned int intel_dp_unused_lane_mask(int lane_count) bool intel_dp_read_dpcd(struct intel_dp *intel_dp); int intel_dp_link_required(int pixel_clock, int bpp); int intel_dp_max_data_rate(int max_link_clock, int max_lanes); -bool intel_digital_port_connected(struct drm_i915_private *dev_priv, - struct intel_digital_port *port); +bool intel_digital_port_connected(struct intel_encoder *encoder); /* intel_dp_aux_backlight.c */ int intel_dp_aux_init_backlight_funcs(struct intel_connector *intel_connector); @@ -1760,6 +1760,7 @@ static inline void intel_backlight_device_unregister(struct intel_connector *con /* intel_psr.c */ +#define CAN_PSR(dev_priv) (HAS_PSR(dev_priv) && dev_priv->psr.sink_support) void intel_psr_enable(struct intel_dp *intel_dp, const struct intel_crtc_state *crtc_state); void intel_psr_disable(struct intel_dp *intel_dp, @@ -1932,6 +1933,8 @@ void skl_update_plane(struct intel_plane *plane, const struct intel_plane_state *plane_state); void skl_disable_plane(struct intel_plane *plane, struct intel_crtc *crtc); bool skl_plane_get_hw_state(struct intel_plane *plane); +bool skl_plane_has_ccs(struct drm_i915_private *dev_priv, + enum pipe pipe, enum plane_id plane_id); /* intel_tv.c */ void intel_tv_init(struct drm_i915_private *dev_priv); diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c index d790bdc227ff..7eebfbb95e89 100644 --- a/drivers/gpu/drm/i915/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/intel_engine_cs.c @@ -38,9 +38,11 @@ */ #define HSW_CXT_TOTAL_SIZE (17 * PAGE_SIZE) +#define DEFAULT_LR_CONTEXT_RENDER_SIZE (22 * PAGE_SIZE) #define GEN8_LR_CONTEXT_RENDER_SIZE (20 * PAGE_SIZE) #define GEN9_LR_CONTEXT_RENDER_SIZE (22 * PAGE_SIZE) #define GEN10_LR_CONTEXT_RENDER_SIZE (18 * PAGE_SIZE) +#define GEN11_LR_CONTEXT_RENDER_SIZE (14 * PAGE_SIZE) #define GEN8_LR_CONTEXT_OTHER_SIZE ( 2 * PAGE_SIZE) @@ -157,6 +159,9 @@ __intel_engine_context_size(struct drm_i915_private *dev_priv, u8 class) switch (INTEL_GEN(dev_priv)) { default: MISSING_CASE(INTEL_GEN(dev_priv)); + return DEFAULT_LR_CONTEXT_RENDER_SIZE; + case 11: + return GEN11_LR_CONTEXT_RENDER_SIZE; case 10: return GEN10_LR_CONTEXT_RENDER_SIZE; case 9: @@ -1389,7 +1394,8 @@ int init_workarounds_ring(struct intel_engine_cs *engine) struct drm_i915_private *dev_priv = engine->i915; int err; - WARN_ON(engine->id != RCS); + if (GEM_WARN_ON(engine->id != RCS)) + return -EINVAL; dev_priv->workarounds.count = 0; dev_priv->workarounds.hw_whitelist_count[engine->id] = 0; @@ -1943,16 +1949,22 @@ intel_engine_lookup_user(struct drm_i915_private *i915, u8 class, u8 instance) */ int intel_enable_engine_stats(struct intel_engine_cs *engine) { + struct intel_engine_execlists *execlists = &engine->execlists; unsigned long flags; + int err = 0; if (!intel_engine_supports_stats(engine)) return -ENODEV; + tasklet_disable(&execlists->tasklet); spin_lock_irqsave(&engine->stats.lock, flags); - if (engine->stats.enabled == ~0) - goto busy; + + if (unlikely(engine->stats.enabled == ~0)) { + err = -EBUSY; + goto unlock; + } + if (engine->stats.enabled++ == 0) { - struct intel_engine_execlists *execlists = &engine->execlists; const struct execlist_port *port = execlists->port; unsigned int num_ports = execlists_num_ports(execlists); @@ -1967,14 +1979,12 @@ int intel_enable_engine_stats(struct intel_engine_cs *engine) if (engine->stats.active) engine->stats.start = engine->stats.enabled_at; } - spin_unlock_irqrestore(&engine->stats.lock, flags); - - return 0; -busy: +unlock: spin_unlock_irqrestore(&engine->stats.lock, flags); + tasklet_enable(&execlists->tasklet); - return -EBUSY; + return err; } static ktime_t __intel_engine_get_busy_time(struct intel_engine_cs *engine) diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c index f88c1b5dae4c..80682a418a70 100644 --- a/drivers/gpu/drm/i915/intel_fbc.c +++ b/drivers/gpu/drm/i915/intel_fbc.c @@ -492,7 +492,8 @@ static void intel_fbc_schedule_activation(struct intel_crtc *crtc) schedule_work(&work->work); } -static void intel_fbc_deactivate(struct drm_i915_private *dev_priv) +static void intel_fbc_deactivate(struct drm_i915_private *dev_priv, + const char *reason) { struct intel_fbc *fbc = &dev_priv->fbc; @@ -505,6 +506,8 @@ static void intel_fbc_deactivate(struct drm_i915_private *dev_priv) if (fbc->active) intel_fbc_hw_deactivate(dev_priv); + + fbc->no_fbc_reason = reason; } static bool multiple_pipes_ok(struct intel_crtc *crtc, @@ -668,11 +671,13 @@ void intel_fbc_cleanup_cfb(struct drm_i915_private *dev_priv) static bool stride_is_valid(struct drm_i915_private *dev_priv, unsigned int stride) { - /* These should have been caught earlier. */ - WARN_ON(stride < 512); - WARN_ON((stride & (64 - 1)) != 0); + /* This should have been caught earlier. */ + if (WARN_ON_ONCE((stride & (64 - 1)) != 0)) + return false; /* Below are the additional FBC restrictions. */ + if (stride < 512) + return false; if (IS_GEN2(dev_priv) || IS_GEN3(dev_priv)) return stride == 4096 || stride == 8192; @@ -921,6 +926,7 @@ void intel_fbc_pre_update(struct intel_crtc *crtc, { struct drm_i915_private *dev_priv = to_i915(crtc->base.dev); struct intel_fbc *fbc = &dev_priv->fbc; + const char *reason = "update pending"; if (!fbc_supported(dev_priv)) return; @@ -928,7 +934,7 @@ void intel_fbc_pre_update(struct intel_crtc *crtc, mutex_lock(&fbc->lock); if (!multiple_pipes_ok(crtc, plane_state)) { - fbc->no_fbc_reason = "more than one pipe active"; + reason = "more than one pipe active"; goto deactivate; } @@ -938,7 +944,7 @@ void intel_fbc_pre_update(struct intel_crtc *crtc, intel_fbc_update_state_cache(crtc, crtc_state, plane_state); deactivate: - intel_fbc_deactivate(dev_priv); + intel_fbc_deactivate(dev_priv, reason); unlock: mutex_unlock(&fbc->lock); } @@ -971,9 +977,8 @@ static void __intel_fbc_post_update(struct intel_crtc *crtc) intel_fbc_reg_params_equal(&old_params, &fbc->params)) return; - intel_fbc_deactivate(dev_priv); + intel_fbc_deactivate(dev_priv, "FBC enabled (active or scheduled)"); intel_fbc_schedule_activation(crtc); - fbc->no_fbc_reason = "FBC enabled (active or scheduled)"; } void intel_fbc_post_update(struct intel_crtc *crtc) @@ -1014,7 +1019,7 @@ void intel_fbc_invalidate(struct drm_i915_private *dev_priv, fbc->busy_bits |= intel_fbc_get_frontbuffer_bit(fbc) & frontbuffer_bits; if (fbc->enabled && fbc->busy_bits) - intel_fbc_deactivate(dev_priv); + intel_fbc_deactivate(dev_priv, "frontbuffer write"); mutex_unlock(&fbc->lock); } @@ -1244,7 +1249,7 @@ static void intel_fbc_underrun_work_fn(struct work_struct *work) DRM_DEBUG_KMS("Disabling FBC due to FIFO underrun.\n"); fbc->underrun_detected = true; - intel_fbc_deactivate(dev_priv); + intel_fbc_deactivate(dev_priv, "FIFO underrun"); out: mutex_unlock(&fbc->lock); } @@ -1371,7 +1376,7 @@ void intel_fbc_init(struct drm_i915_private *dev_priv) for_each_pipe(dev_priv, pipe) { fbc->possible_framebuffer_bits |= - INTEL_FRONTBUFFER_PRIMARY(pipe); + INTEL_FRONTBUFFER(pipe, PLANE_PRIMARY); if (fbc_on_pipe_a_only(dev_priv)) break; diff --git a/drivers/gpu/drm/i915/intel_guc.c b/drivers/gpu/drm/i915/intel_guc.c index 3c6bf5a34c3c..21140ccd7a97 100644 --- a/drivers/gpu/drm/i915/intel_guc.c +++ b/drivers/gpu/drm/i915/intel_guc.c @@ -23,6 +23,7 @@ */ #include "intel_guc.h" +#include "intel_guc_ads.h" #include "intel_guc_submission.h" #include "i915_drv.h" @@ -63,6 +64,7 @@ void intel_guc_init_early(struct intel_guc *guc) { intel_guc_fw_init_early(guc); intel_guc_ct_init_early(&guc->ct); + intel_guc_log_init_early(guc); mutex_init(&guc->send_mutex); guc->send = intel_guc_send_nop; @@ -86,8 +88,10 @@ int intel_guc_init_wq(struct intel_guc *guc) */ guc->log.runtime.flush_wq = alloc_ordered_workqueue("i915-guc_log", WQ_HIGHPRI | WQ_FREEZABLE); - if (!guc->log.runtime.flush_wq) + if (!guc->log.runtime.flush_wq) { + DRM_ERROR("Couldn't allocate workqueue for GuC log\n"); return -ENOMEM; + } /* * Even though both sending GuC action, and adding a new workitem to @@ -108,6 +112,8 @@ int intel_guc_init_wq(struct intel_guc *guc) WQ_HIGHPRI); if (!guc->preempt_wq) { destroy_workqueue(guc->log.runtime.flush_wq); + DRM_ERROR("Couldn't allocate workqueue for GuC " + "preemption\n"); return -ENOMEM; } } @@ -163,10 +169,25 @@ int intel_guc_init(struct intel_guc *guc) return ret; GEM_BUG_ON(!guc->shared_data); + ret = intel_guc_log_create(guc); + if (ret) + goto err_shared; + + ret = intel_guc_ads_create(guc); + if (ret) + goto err_log; + GEM_BUG_ON(!guc->ads_vma); + /* We need to notify the guc whenever we change the GGTT */ i915_ggtt_enable_guc(dev_priv); return 0; + +err_log: + intel_guc_log_destroy(guc); +err_shared: + guc_shared_data_destroy(guc); + return ret; } void intel_guc_fini(struct intel_guc *guc) @@ -174,6 +195,8 @@ void intel_guc_fini(struct intel_guc *guc) struct drm_i915_private *dev_priv = guc_to_i915(guc); i915_ggtt_disable_guc(dev_priv); + intel_guc_ads_destroy(guc); + intel_guc_log_destroy(guc); guc_shared_data_destroy(guc); } @@ -197,6 +220,19 @@ static u32 get_core_family(struct drm_i915_private *dev_priv) } } +static u32 get_log_verbosity_flags(void) +{ + if (i915_modparams.guc_log_level > 0) { + u32 verbosity = i915_modparams.guc_log_level - 1; + + GEM_BUG_ON(verbosity > GUC_LOG_VERBOSITY_MAX); + return verbosity << GUC_LOG_VERBOSITY_SHIFT; + } + + GEM_BUG_ON(i915_modparams.enable_guc < 0); + return GUC_LOG_DISABLED; +} + /* * Initialise the GuC parameter block before starting the firmware * transfer. These parameters are read by the firmware on startup @@ -229,12 +265,7 @@ void intel_guc_init_params(struct intel_guc *guc) params[GUC_CTL_LOG_PARAMS] = guc->log.flags; - if (i915_modparams.guc_log_level >= 0) { - params[GUC_CTL_DEBUG] = - i915_modparams.guc_log_level << GUC_LOG_VERBOSITY_SHIFT; - } else { - params[GUC_CTL_DEBUG] = GUC_LOG_DISABLED; - } + params[GUC_CTL_DEBUG] = get_log_verbosity_flags(); /* If GuC submission is enabled, set up additional parameters here */ if (USES_GUC_SUBMISSION(dev_priv)) { @@ -427,7 +458,7 @@ int intel_guc_resume(struct drm_i915_private *dev_priv) if (guc->fw.load_status != INTEL_UC_FIRMWARE_SUCCESS) return 0; - if (i915_modparams.guc_log_level >= 0) + if (i915_modparams.guc_log_level) gen9_enable_guc_interrupts(dev_priv); data[0] = INTEL_GUC_ACTION_EXIT_S_STATE; diff --git a/drivers/gpu/drm/i915/intel_guc_ads.c b/drivers/gpu/drm/i915/intel_guc_ads.c new file mode 100644 index 000000000000..ac627534667d --- /dev/null +++ b/drivers/gpu/drm/i915/intel_guc_ads.c @@ -0,0 +1,151 @@ +/* + * Copyright © 2014-2017 Intel Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#include "intel_guc_ads.h" +#include "intel_uc.h" +#include "i915_drv.h" + +/* + * The Additional Data Struct (ADS) has pointers for different buffers used by + * the GuC. One single gem object contains the ADS struct itself (guc_ads), the + * scheduling policies (guc_policies), a structure describing a collection of + * register sets (guc_mmio_reg_state) and some extra pages for the GuC to save + * its internal state for sleep. + */ + +static void guc_policy_init(struct guc_policy *policy) +{ + policy->execution_quantum = POLICY_DEFAULT_EXECUTION_QUANTUM_US; + policy->preemption_time = POLICY_DEFAULT_PREEMPTION_TIME_US; + policy->fault_time = POLICY_DEFAULT_FAULT_TIME_US; + policy->policy_flags = 0; +} + +static void guc_policies_init(struct guc_policies *policies) +{ + struct guc_policy *policy; + u32 p, i; + + policies->dpc_promote_time = POLICY_DEFAULT_DPC_PROMOTE_TIME_US; + policies->max_num_work_items = POLICY_MAX_NUM_WI; + + for (p = 0; p < GUC_CLIENT_PRIORITY_NUM; p++) { + for (i = GUC_RENDER_ENGINE; i < GUC_MAX_ENGINES_NUM; i++) { + policy = &policies->policy[p][i]; + + guc_policy_init(policy); + } + } + + policies->is_valid = 1; +} + +/* + * The first 80 dwords of the register state context, containing the + * execlists and ppgtt registers. + */ +#define LR_HW_CONTEXT_SIZE (80 * sizeof(u32)) + +/** + * intel_guc_ads_create() - creates GuC ADS + * @guc: intel_guc struct + * + */ +int intel_guc_ads_create(struct intel_guc *guc) +{ + struct drm_i915_private *dev_priv = guc_to_i915(guc); + struct i915_vma *vma; + struct page *page; + /* The ads obj includes the struct itself and buffers passed to GuC */ + struct { + struct guc_ads ads; + struct guc_policies policies; + struct guc_mmio_reg_state reg_state; + u8 reg_state_buffer[GUC_S3_SAVE_SPACE_PAGES * PAGE_SIZE]; + } __packed *blob; + struct intel_engine_cs *engine; + enum intel_engine_id id; + const u32 skipped_offset = LRC_HEADER_PAGES * PAGE_SIZE; + const u32 skipped_size = LRC_PPHWSP_SZ * PAGE_SIZE + LR_HW_CONTEXT_SIZE; + u32 base; + + GEM_BUG_ON(guc->ads_vma); + + vma = intel_guc_allocate_vma(guc, PAGE_ALIGN(sizeof(*blob))); + if (IS_ERR(vma)) + return PTR_ERR(vma); + + guc->ads_vma = vma; + + page = i915_vma_first_page(vma); + blob = kmap(page); + + /* GuC scheduling policies */ + guc_policies_init(&blob->policies); + + /* MMIO reg state */ + for_each_engine(engine, dev_priv, id) { + blob->reg_state.white_list[engine->guc_id].mmio_start = + engine->mmio_base + GUC_MMIO_WHITE_LIST_START; + + /* Nothing to be saved or restored for now. */ + blob->reg_state.white_list[engine->guc_id].count = 0; + } + + /* + * The GuC requires a "Golden Context" when it reinitialises + * engines after a reset. Here we use the Render ring default + * context, which must already exist and be pinned in the GGTT, + * so its address won't change after we've told the GuC where + * to find it. Note that we have to skip our header (1 page), + * because our GuC shared data is there. + */ + blob->ads.golden_context_lrca = + guc_ggtt_offset(dev_priv->kernel_context->engine[RCS].state) + + skipped_offset; + + /* + * The GuC expects us to exclude the portion of the context image that + * it skips from the size it is to read. It starts reading from after + * the execlist context (so skipping the first page [PPHWSP] and 80 + * dwords). Weird guc is weird. + */ + for_each_engine(engine, dev_priv, id) + blob->ads.eng_state_size[engine->guc_id] = + engine->context_size - skipped_size; + + base = guc_ggtt_offset(vma); + blob->ads.scheduler_policies = base + ptr_offset(blob, policies); + blob->ads.reg_state_buffer = base + ptr_offset(blob, reg_state_buffer); + blob->ads.reg_state_addr = base + ptr_offset(blob, reg_state); + + kunmap(page); + + return 0; +} + +void intel_guc_ads_destroy(struct intel_guc *guc) +{ + i915_vma_unpin_and_release(&guc->ads_vma); +} diff --git a/drivers/gpu/drm/i915/intel_guc_ads.h b/drivers/gpu/drm/i915/intel_guc_ads.h new file mode 100644 index 000000000000..c4735742c564 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_guc_ads.h @@ -0,0 +1,33 @@ +/* + * Copyright © 2014-2017 Intel Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + */ + +#ifndef _INTEL_GUC_ADS_H_ +#define _INTEL_GUC_ADS_H_ + +struct intel_guc; + +int intel_guc_ads_create(struct intel_guc *guc); +void intel_guc_ads_destroy(struct intel_guc *guc); + +#endif diff --git a/drivers/gpu/drm/i915/intel_guc_log.c b/drivers/gpu/drm/i915/intel_guc_log.c index eaedd63e3819..7b5074e2120c 100644 --- a/drivers/gpu/drm/i915/intel_guc_log.c +++ b/drivers/gpu/drm/i915/intel_guc_log.c @@ -33,11 +33,10 @@ static void guc_log_capture_logs(struct intel_guc *guc); /** * DOC: GuC firmware log * - * Firmware log is enabled by setting i915.guc_log_level to non-negative level. + * Firmware log is enabled by setting i915.guc_log_level to the positive level. * Log data is printed out via reading debugfs i915_guc_log_dump. Reading from * i915_guc_load_status will print out firmware loading status and scratch * registers value. - * */ static int guc_log_flush_complete(struct intel_guc *guc) @@ -59,11 +58,15 @@ static int guc_log_flush(struct intel_guc *guc) return intel_guc_send(guc, action, ARRAY_SIZE(action)); } -static int guc_log_control(struct intel_guc *guc, u32 control_val) +static int guc_log_control(struct intel_guc *guc, bool enable, u32 verbosity) { + union guc_log_control control_val = { + .logging_enabled = enable, + .verbosity = verbosity, + }; u32 action[] = { INTEL_GUC_ACTION_UK_LOG_ENABLE_LOGGING, - control_val + control_val.value }; return intel_guc_send(guc, action, ARRAY_SIZE(action)); @@ -78,7 +81,8 @@ static int subbuf_start_callback(struct rchan_buf *buf, void *prev_subbuf, size_t prev_padding) { - /* Use no-overwrite mode by default, where relay will stop accepting + /* + * Use no-overwrite mode by default, where relay will stop accepting * new data if there are no empty sub buffers left. * There is no strict synchronization enforced by relay between Consumer * and Producer. In overwrite mode, there is a possibility of getting @@ -104,7 +108,8 @@ static struct dentry *create_buf_file_callback(const char *filename, { struct dentry *buf_file; - /* This to enable the use of a single buffer for the relay channel and + /* + * This to enable the use of a single buffer for the relay channel and * correspondingly have a single file exposed to User, through which * it can collect the logs in order without any post-processing. * Need to set 'is_global' even if parent is NULL for early logging. @@ -114,7 +119,8 @@ static struct dentry *create_buf_file_callback(const char *filename, if (!parent) return NULL; - /* Not using the channel filename passed as an argument, since for each + /* + * Not using the channel filename passed as an argument, since for each * channel relay appends the corresponding CPU number to the filename * passed in relay_open(). This should be fine as relay just needs a * dentry of the file associated with the channel buffer and that file's @@ -147,13 +153,16 @@ static int guc_log_relay_file_create(struct intel_guc *guc) struct dentry *log_dir; int ret; - if (i915_modparams.guc_log_level < 0) + if (!i915_modparams.guc_log_level) return 0; + mutex_lock(&guc->log.runtime.relay_lock); + /* For now create the log file in /sys/kernel/debug/dri/0 dir */ log_dir = dev_priv->drm.primary->debugfs_root; - /* If /sys/kernel/debug/dri/0 location do not exist, then debugfs is + /* + * If /sys/kernel/debug/dri/0 location do not exist, then debugfs is * not mounted and so can't create the relay file. * The relay API seems to fit well with debugfs only, for availing relay * there are 3 requirements which can be met for debugfs file only in a @@ -166,25 +175,41 @@ static int guc_log_relay_file_create(struct intel_guc *guc) */ if (!log_dir) { DRM_ERROR("Debugfs dir not available yet for GuC log file\n"); - return -ENODEV; + ret = -ENODEV; + goto out_unlock; } ret = relay_late_setup_files(guc->log.runtime.relay_chan, "guc_log", log_dir); if (ret < 0 && ret != -EEXIST) { DRM_ERROR("Couldn't associate relay chan with file %d\n", ret); - return ret; + goto out_unlock; } - return 0; + ret = 0; + +out_unlock: + mutex_unlock(&guc->log.runtime.relay_lock); + return ret; +} + +static bool guc_log_has_relay(struct intel_guc *guc) +{ + lockdep_assert_held(&guc->log.runtime.relay_lock); + + return guc->log.runtime.relay_chan != NULL; } static void guc_move_to_next_buf(struct intel_guc *guc) { - /* Make sure the updates made in the sub buffer are visible when + /* + * Make sure the updates made in the sub buffer are visible when * Consumer sees the following update to offset inside the sub buffer. */ smp_wmb(); + if (!guc_log_has_relay(guc)) + return; + /* All data has been written, so now move the offset of sub buffer. */ relay_reserve(guc->log.runtime.relay_chan, guc->log.vma->obj->base.size); @@ -194,10 +219,11 @@ static void guc_move_to_next_buf(struct intel_guc *guc) static void *guc_get_write_buffer(struct intel_guc *guc) { - if (!guc->log.runtime.relay_chan) + if (!guc_log_has_relay(guc)) return NULL; - /* Just get the base address of a new sub buffer and copy data into it + /* + * Just get the base address of a new sub buffer and copy data into it * ourselves. NULL will be returned in no-overwrite mode, if all sub * buffers are full. Could have used the relay_write() to indirectly * copy the data, but that would have been bit convoluted, as we need to @@ -262,15 +288,30 @@ static void guc_read_update_log_buffer(struct intel_guc *guc) /* Get the pointer to shared GuC log buffer */ log_buf_state = src_data = guc->log.runtime.buf_addr; + mutex_lock(&guc->log.runtime.relay_lock); + /* Get the pointer to local buffer to store the logs */ log_buf_snapshot_state = dst_data = guc_get_write_buffer(guc); + if (unlikely(!log_buf_snapshot_state)) { + /* + * Used rate limited to avoid deluge of messages, logs might be + * getting consumed by User at a slow rate. + */ + DRM_ERROR_RATELIMITED("no sub-buffer to capture logs\n"); + guc->log.capture_miss_count++; + mutex_unlock(&guc->log.runtime.relay_lock); + + return; + } + /* Actual logs are present from the 2nd page */ src_data += PAGE_SIZE; dst_data += PAGE_SIZE; for (type = GUC_ISR_LOG_BUFFER; type < GUC_MAX_LOG_BUFFER; type++) { - /* Make a copy of the state structure, inside GuC log buffer + /* + * Make a copy of the state structure, inside GuC log buffer * (which is uncached mapped), on the stack to avoid reading * from it multiple times. */ @@ -290,14 +331,12 @@ static void guc_read_update_log_buffer(struct intel_guc *guc) log_buf_state->flush_to_file = 0; log_buf_state++; - if (unlikely(!log_buf_snapshot_state)) - continue; - /* First copy the state structure in snapshot buffer */ memcpy(log_buf_snapshot_state, &log_buf_state_local, sizeof(struct guc_log_buffer_state)); - /* The write pointer could have been updated by GuC firmware, + /* + * The write pointer could have been updated by GuC firmware, * after sending the flush interrupt to Host, for consistency * set write pointer value to same value of sampled_write_ptr * in the snapshot buffer. @@ -332,15 +371,9 @@ static void guc_read_update_log_buffer(struct intel_guc *guc) dst_data += buffer_size; } - if (log_buf_snapshot_state) - guc_move_to_next_buf(guc); - else { - /* Used rate limited to avoid deluge of messages, logs might be - * getting consumed by User at a slow rate. - */ - DRM_ERROR_RATELIMITED("no sub-buffer to capture logs\n"); - guc->log.capture_miss_count++; - } + guc_move_to_next_buf(guc); + + mutex_unlock(&guc->log.runtime.relay_lock); } static void capture_logs_work(struct work_struct *work) @@ -360,19 +393,21 @@ static int guc_log_runtime_create(struct intel_guc *guc) { struct drm_i915_private *dev_priv = guc_to_i915(guc); void *vaddr; - struct rchan *guc_log_relay_chan; - size_t n_subbufs, subbuf_size; int ret; lockdep_assert_held(&dev_priv->drm.struct_mutex); + if (!guc->log.vma) + return -ENODEV; + GEM_BUG_ON(guc_log_has_runtime(guc)); ret = i915_gem_object_set_to_wc_domain(guc->log.vma->obj, true); if (ret) return ret; - /* Create a WC (Uncached for read) vmalloc mapping of log + /* + * Create a WC (Uncached for read) vmalloc mapping of log * buffer pages, so that we can directly get the data * (up-to-date) from memory. */ @@ -384,17 +419,55 @@ static int guc_log_runtime_create(struct intel_guc *guc) guc->log.runtime.buf_addr = vaddr; + return 0; +} + +static void guc_log_runtime_destroy(struct intel_guc *guc) +{ + /* + * It's possible that the runtime stuff was never allocated because + * GuC log was disabled at the boot time. + */ + if (!guc_log_has_runtime(guc)) + return; + + i915_gem_object_unpin_map(guc->log.vma->obj); + guc->log.runtime.buf_addr = NULL; +} + +void intel_guc_log_init_early(struct intel_guc *guc) +{ + mutex_init(&guc->log.runtime.relay_lock); + INIT_WORK(&guc->log.runtime.flush_work, capture_logs_work); +} + +int intel_guc_log_relay_create(struct intel_guc *guc) +{ + struct drm_i915_private *dev_priv = guc_to_i915(guc); + struct rchan *guc_log_relay_chan; + size_t n_subbufs, subbuf_size; + int ret; + + if (!i915_modparams.guc_log_level) + return 0; + + mutex_lock(&guc->log.runtime.relay_lock); + + GEM_BUG_ON(guc_log_has_relay(guc)); + /* Keep the size of sub buffers same as shared log buffer */ - subbuf_size = guc->log.vma->obj->base.size; + subbuf_size = GUC_LOG_SIZE; - /* Store up to 8 snapshots, which is large enough to buffer sufficient + /* + * Store up to 8 snapshots, which is large enough to buffer sufficient * boot time logs and provides enough leeway to User, in terms of * latency, for consuming the logs from relay. Also doesn't take * up too much memory. */ n_subbufs = 8; - /* Create a relay channel, so that we have buffers for storing + /* + * Create a relay channel, so that we have buffers for storing * the GuC firmware logs, the channel will be linked with a file * later on when debugfs is registered. */ @@ -404,33 +477,39 @@ static int guc_log_runtime_create(struct intel_guc *guc) DRM_ERROR("Couldn't create relay chan for GuC logging\n"); ret = -ENOMEM; - goto err_vaddr; + goto err; } GEM_BUG_ON(guc_log_relay_chan->subbuf_size < subbuf_size); guc->log.runtime.relay_chan = guc_log_relay_chan; - INIT_WORK(&guc->log.runtime.flush_work, capture_logs_work); + mutex_unlock(&guc->log.runtime.relay_lock); + return 0; -err_vaddr: - i915_gem_object_unpin_map(guc->log.vma->obj); - guc->log.runtime.buf_addr = NULL; +err: + mutex_unlock(&guc->log.runtime.relay_lock); + /* logging will be off */ + i915_modparams.guc_log_level = 0; return ret; } -static void guc_log_runtime_destroy(struct intel_guc *guc) +void intel_guc_log_relay_destroy(struct intel_guc *guc) { + mutex_lock(&guc->log.runtime.relay_lock); + /* - * It's possible that the runtime stuff was never allocated because - * guc_log_level was < 0 at the time - **/ - if (!guc_log_has_runtime(guc)) - return; + * It's possible that the relay was never allocated because + * GuC log was disabled at the boot time. + */ + if (!guc_log_has_relay(guc)) + goto out_unlock; relay_close(guc->log.runtime.relay_chan); - i915_gem_object_unpin_map(guc->log.vma->obj); - guc->log.runtime.buf_addr = NULL; + guc->log.runtime.relay_chan = NULL; + +out_unlock: + mutex_unlock(&guc->log.runtime.relay_lock); } static int guc_log_late_setup(struct intel_guc *guc) @@ -438,16 +517,24 @@ static int guc_log_late_setup(struct intel_guc *guc) struct drm_i915_private *dev_priv = guc_to_i915(guc); int ret; - lockdep_assert_held(&dev_priv->drm.struct_mutex); - if (!guc_log_has_runtime(guc)) { - /* If log_level was set as -1 at boot time, then setup needed to - * handle log buffer flush interrupts would not have been done yet, - * so do that now. + /* + * If log was disabled at boot time, then setup needed to handle + * log buffer flush interrupts would not have been done yet, so + * do that now. */ - ret = guc_log_runtime_create(guc); + ret = intel_guc_log_relay_create(guc); if (ret) goto err; + + mutex_lock(&dev_priv->drm.struct_mutex); + intel_runtime_pm_get(dev_priv); + ret = guc_log_runtime_create(guc); + intel_runtime_pm_put(dev_priv); + mutex_unlock(&dev_priv->drm.struct_mutex); + + if (ret) + goto err_relay; } ret = guc_log_relay_file_create(guc); @@ -457,10 +544,14 @@ static int guc_log_late_setup(struct intel_guc *guc) return 0; err_runtime: + mutex_lock(&dev_priv->drm.struct_mutex); guc_log_runtime_destroy(guc); + mutex_unlock(&dev_priv->drm.struct_mutex); +err_relay: + intel_guc_log_relay_destroy(guc); err: /* logging will remain off */ - i915_modparams.guc_log_level = -1; + i915_modparams.guc_log_level = 0; return ret; } @@ -470,7 +561,8 @@ static void guc_log_capture_logs(struct intel_guc *guc) guc_read_update_log_buffer(guc); - /* Generally device is expected to be active only at this + /* + * Generally device is expected to be active only at this * time, so get/put should be really quick. */ intel_runtime_pm_get(dev_priv); @@ -482,20 +574,26 @@ static void guc_flush_logs(struct intel_guc *guc) { struct drm_i915_private *dev_priv = guc_to_i915(guc); - if (!USES_GUC_SUBMISSION(dev_priv) || - (i915_modparams.guc_log_level < 0)) + if (!USES_GUC_SUBMISSION(dev_priv) || !i915_modparams.guc_log_level) return; /* First disable the interrupts, will be renabled afterwards */ + mutex_lock(&dev_priv->drm.struct_mutex); + intel_runtime_pm_get(dev_priv); gen9_disable_guc_interrupts(dev_priv); + intel_runtime_pm_put(dev_priv); + mutex_unlock(&dev_priv->drm.struct_mutex); - /* Before initiating the forceful flush, wait for any pending/ongoing + /* + * Before initiating the forceful flush, wait for any pending/ongoing * flush to complete otherwise forceful flush may not actually happen. */ flush_work(&guc->log.runtime.flush_work); /* Ask GuC to update the log buffer state */ + intel_runtime_pm_get(dev_priv); guc_log_flush(guc); + intel_runtime_pm_put(dev_priv); /* GuC would have updated log buffer by now, so capture it */ guc_log_capture_logs(guc); @@ -506,21 +604,12 @@ int intel_guc_log_create(struct intel_guc *guc) struct i915_vma *vma; unsigned long offset; u32 flags; - u32 size; int ret; GEM_BUG_ON(guc->log.vma); - if (i915_modparams.guc_log_level > GUC_LOG_VERBOSITY_MAX) - i915_modparams.guc_log_level = GUC_LOG_VERBOSITY_MAX; - - /* The first page is to save log buffer state. Allocate one - * extra page for others in case for overlap */ - size = (1 + GUC_LOG_DPC_PAGES + 1 + - GUC_LOG_ISR_PAGES + 1 + - GUC_LOG_CRASH_PAGES + 1) << PAGE_SHIFT; - - /* We require SSE 4.1 for fast reads from the GuC log buffer and + /* + * We require SSE 4.1 for fast reads from the GuC log buffer and * it should be present on the chipsets supporting GuC based * submisssions. */ @@ -529,7 +618,7 @@ int intel_guc_log_create(struct intel_guc *guc) goto err; } - vma = intel_guc_allocate_vma(guc, size); + vma = intel_guc_allocate_vma(guc, GUC_LOG_SIZE); if (IS_ERR(vma)) { ret = PTR_ERR(vma); goto err; @@ -537,7 +626,7 @@ int intel_guc_log_create(struct intel_guc *guc) guc->log.vma = vma; - if (i915_modparams.guc_log_level >= 0) { + if (i915_modparams.guc_log_level) { ret = guc_log_runtime_create(guc); if (ret < 0) goto err_vma; @@ -558,7 +647,7 @@ err_vma: i915_vma_unpin_and_release(&guc->log.vma); err: /* logging will be off */ - i915_modparams.guc_log_level = -1; + i915_modparams.guc_log_level = 0; return ret; } @@ -568,35 +657,46 @@ void intel_guc_log_destroy(struct intel_guc *guc) i915_vma_unpin_and_release(&guc->log.vma); } -int i915_guc_log_control(struct drm_i915_private *dev_priv, u64 control_val) +int intel_guc_log_control(struct intel_guc *guc, u64 control_val) { - struct intel_guc *guc = &dev_priv->guc; - - union guc_log_control log_param; + struct drm_i915_private *dev_priv = guc_to_i915(guc); + bool enable_logging = control_val > 0; + u32 verbosity; int ret; - log_param.value = control_val; + if (!guc->log.vma) + return -ENODEV; - if (log_param.verbosity < GUC_LOG_VERBOSITY_MIN || - log_param.verbosity > GUC_LOG_VERBOSITY_MAX) + BUILD_BUG_ON(GUC_LOG_VERBOSITY_MIN); + if (control_val > 1 + GUC_LOG_VERBOSITY_MAX) return -EINVAL; /* This combination doesn't make sense & won't have any effect */ - if (!log_param.logging_enabled && (i915_modparams.guc_log_level < 0)) + if (!enable_logging && !i915_modparams.guc_log_level) return 0; - ret = guc_log_control(guc, log_param.value); + verbosity = enable_logging ? control_val - 1 : 0; + + ret = mutex_lock_interruptible(&dev_priv->drm.struct_mutex); + if (ret) + return ret; + intel_runtime_pm_get(dev_priv); + ret = guc_log_control(guc, enable_logging, verbosity); + intel_runtime_pm_put(dev_priv); + mutex_unlock(&dev_priv->drm.struct_mutex); + if (ret < 0) { DRM_DEBUG_DRIVER("guc_logging_control action failed %d\n", ret); return ret; } - if (log_param.logging_enabled) { - i915_modparams.guc_log_level = log_param.verbosity; + if (enable_logging) { + i915_modparams.guc_log_level = 1 + verbosity; - /* If log_level was set as -1 at boot time, then the relay channel file - * wouldn't have been created by now and interrupts also would not have - * been enabled. Try again now, just in case. + /* + * If log was disabled at boot time, then the relay channel file + * wouldn't have been created by now and interrupts also would + * not have been enabled. Try again now, just in case. */ ret = guc_log_late_setup(guc); if (ret < 0) { @@ -605,9 +705,14 @@ int i915_guc_log_control(struct drm_i915_private *dev_priv, u64 control_val) } /* GuC logging is currently the only user of Guc2Host interrupts */ + mutex_lock(&dev_priv->drm.struct_mutex); + intel_runtime_pm_get(dev_priv); gen9_enable_guc_interrupts(dev_priv); + intel_runtime_pm_put(dev_priv); + mutex_unlock(&dev_priv->drm.struct_mutex); } else { - /* Once logging is disabled, GuC won't generate logs & send an + /* + * Once logging is disabled, GuC won't generate logs & send an * interrupt. But there could be some data in the log buffer * which is yet to be captured. So request GuC to update the log * buffer state and then collect the left over logs. @@ -615,7 +720,7 @@ int i915_guc_log_control(struct drm_i915_private *dev_priv, u64 control_val) guc_flush_logs(guc); /* As logging is disabled, update log level to reflect that */ - i915_modparams.guc_log_level = -1; + i915_modparams.guc_log_level = 0; } return ret; @@ -623,23 +728,27 @@ int i915_guc_log_control(struct drm_i915_private *dev_priv, u64 control_val) void i915_guc_log_register(struct drm_i915_private *dev_priv) { - if (!USES_GUC_SUBMISSION(dev_priv) || - (i915_modparams.guc_log_level < 0)) + if (!USES_GUC_SUBMISSION(dev_priv) || !i915_modparams.guc_log_level) return; - mutex_lock(&dev_priv->drm.struct_mutex); guc_log_late_setup(&dev_priv->guc); - mutex_unlock(&dev_priv->drm.struct_mutex); } void i915_guc_log_unregister(struct drm_i915_private *dev_priv) { + struct intel_guc *guc = &dev_priv->guc; + if (!USES_GUC_SUBMISSION(dev_priv)) return; mutex_lock(&dev_priv->drm.struct_mutex); /* GuC logging is currently the only user of Guc2Host interrupts */ + intel_runtime_pm_get(dev_priv); gen9_disable_guc_interrupts(dev_priv); - guc_log_runtime_destroy(&dev_priv->guc); + intel_runtime_pm_put(dev_priv); + + guc_log_runtime_destroy(guc); mutex_unlock(&dev_priv->drm.struct_mutex); + + intel_guc_log_relay_destroy(guc); } diff --git a/drivers/gpu/drm/i915/intel_guc_log.h b/drivers/gpu/drm/i915/intel_guc_log.h index f512cf79339b..dab0e949567a 100644 --- a/drivers/gpu/drm/i915/intel_guc_log.h +++ b/drivers/gpu/drm/i915/intel_guc_log.h @@ -32,6 +32,13 @@ struct drm_i915_private; struct intel_guc; +/* + * The first page is to save log buffer state. Allocate one + * extra page for others in case for overlap + */ +#define GUC_LOG_SIZE ((1 + GUC_LOG_DPC_PAGES + 1 + GUC_LOG_ISR_PAGES + \ + 1 + GUC_LOG_CRASH_PAGES + 1) << PAGE_SHIFT) + struct intel_guc_log { u32 flags; struct i915_vma *vma; @@ -41,6 +48,8 @@ struct intel_guc_log { struct workqueue_struct *flush_wq; struct work_struct flush_work; struct rchan *relay_chan; + /* To serialize the access to relay_chan */ + struct mutex relay_lock; } runtime; /* logging related stats */ u32 capture_miss_count; @@ -52,7 +61,10 @@ struct intel_guc_log { int intel_guc_log_create(struct intel_guc *guc); void intel_guc_log_destroy(struct intel_guc *guc); -int i915_guc_log_control(struct drm_i915_private *dev_priv, u64 control_val); +void intel_guc_log_init_early(struct intel_guc *guc); +int intel_guc_log_relay_create(struct intel_guc *guc); +void intel_guc_log_relay_destroy(struct intel_guc *guc); +int intel_guc_log_control(struct intel_guc *guc, u64 control_val); void i915_guc_log_register(struct drm_i915_private *dev_priv); void i915_guc_log_unregister(struct drm_i915_private *dev_priv); diff --git a/drivers/gpu/drm/i915/intel_guc_submission.c b/drivers/gpu/drm/i915/intel_guc_submission.c index 4d2409466a3a..1f3a8786bbdc 100644 --- a/drivers/gpu/drm/i915/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/intel_guc_submission.c @@ -73,13 +73,6 @@ * ELSP context descriptor dword into Work Item. * See guc_add_request() * - * ADS: - * The Additional Data Struct (ADS) has pointers for different buffers used by - * the GuC. One single gem object contains the ADS struct itself (guc_ads), the - * scheduling policies (guc_policies), a structure describing a collection of - * register sets (guc_mmio_reg_state) and some extra pages for the GuC to save - * its internal state for sleep. - * */ static inline bool is_high_priority(struct intel_guc_client *client) @@ -1012,117 +1005,6 @@ static void guc_clients_destroy(struct intel_guc *guc) guc_client_free(client); } -static void guc_policy_init(struct guc_policy *policy) -{ - policy->execution_quantum = POLICY_DEFAULT_EXECUTION_QUANTUM_US; - policy->preemption_time = POLICY_DEFAULT_PREEMPTION_TIME_US; - policy->fault_time = POLICY_DEFAULT_FAULT_TIME_US; - policy->policy_flags = 0; -} - -static void guc_policies_init(struct guc_policies *policies) -{ - struct guc_policy *policy; - u32 p, i; - - policies->dpc_promote_time = POLICY_DEFAULT_DPC_PROMOTE_TIME_US; - policies->max_num_work_items = POLICY_MAX_NUM_WI; - - for (p = 0; p < GUC_CLIENT_PRIORITY_NUM; p++) { - for (i = GUC_RENDER_ENGINE; i < GUC_MAX_ENGINES_NUM; i++) { - policy = &policies->policy[p][i]; - - guc_policy_init(policy); - } - } - - policies->is_valid = 1; -} - -/* - * The first 80 dwords of the register state context, containing the - * execlists and ppgtt registers. - */ -#define LR_HW_CONTEXT_SIZE (80 * sizeof(u32)) - -static int guc_ads_create(struct intel_guc *guc) -{ - struct drm_i915_private *dev_priv = guc_to_i915(guc); - struct i915_vma *vma; - struct page *page; - /* The ads obj includes the struct itself and buffers passed to GuC */ - struct { - struct guc_ads ads; - struct guc_policies policies; - struct guc_mmio_reg_state reg_state; - u8 reg_state_buffer[GUC_S3_SAVE_SPACE_PAGES * PAGE_SIZE]; - } __packed *blob; - struct intel_engine_cs *engine; - enum intel_engine_id id; - const u32 skipped_offset = LRC_HEADER_PAGES * PAGE_SIZE; - const u32 skipped_size = LRC_PPHWSP_SZ * PAGE_SIZE + LR_HW_CONTEXT_SIZE; - u32 base; - - GEM_BUG_ON(guc->ads_vma); - - vma = intel_guc_allocate_vma(guc, PAGE_ALIGN(sizeof(*blob))); - if (IS_ERR(vma)) - return PTR_ERR(vma); - - guc->ads_vma = vma; - - page = i915_vma_first_page(vma); - blob = kmap(page); - - /* GuC scheduling policies */ - guc_policies_init(&blob->policies); - - /* MMIO reg state */ - for_each_engine(engine, dev_priv, id) { - blob->reg_state.white_list[engine->guc_id].mmio_start = - engine->mmio_base + GUC_MMIO_WHITE_LIST_START; - - /* Nothing to be saved or restored for now. */ - blob->reg_state.white_list[engine->guc_id].count = 0; - } - - /* - * The GuC requires a "Golden Context" when it reinitialises - * engines after a reset. Here we use the Render ring default - * context, which must already exist and be pinned in the GGTT, - * so its address won't change after we've told the GuC where - * to find it. Note that we have to skip our header (1 page), - * because our GuC shared data is there. - */ - blob->ads.golden_context_lrca = - guc_ggtt_offset(dev_priv->kernel_context->engine[RCS].state) + - skipped_offset; - - /* - * The GuC expects us to exclude the portion of the context image that - * it skips from the size it is to read. It starts reading from after - * the execlist context (so skipping the first page [PPHWSP] and 80 - * dwords). Weird guc is weird. - */ - for_each_engine(engine, dev_priv, id) - blob->ads.eng_state_size[engine->guc_id] = - engine->context_size - skipped_size; - - base = guc_ggtt_offset(vma); - blob->ads.scheduler_policies = base + ptr_offset(blob, policies); - blob->ads.reg_state_buffer = base + ptr_offset(blob, reg_state_buffer); - blob->ads.reg_state_addr = base + ptr_offset(blob, reg_state); - - kunmap(page); - - return 0; -} - -static void guc_ads_destroy(struct intel_guc *guc) -{ - i915_vma_unpin_and_release(&guc->ads_vma); -} - /* * Set up the memory resources to be shared with the GuC (via the GGTT) * at firmware loading time. @@ -1146,15 +1028,6 @@ int intel_guc_submission_init(struct intel_guc *guc) */ GEM_BUG_ON(!guc->stage_desc_pool); - ret = intel_guc_log_create(guc); - if (ret < 0) - goto err_stage_desc_pool; - - ret = guc_ads_create(guc); - if (ret < 0) - goto err_log; - GEM_BUG_ON(!guc->ads_vma); - WARN_ON(!guc_verify_doorbells(guc)); ret = guc_clients_create(guc); if (ret) @@ -1167,11 +1040,6 @@ int intel_guc_submission_init(struct intel_guc *guc) return 0; -err_log: - intel_guc_log_destroy(guc); -err_stage_desc_pool: - guc_stage_desc_pool_destroy(guc); - return ret; } void intel_guc_submission_fini(struct intel_guc *guc) @@ -1186,8 +1054,6 @@ void intel_guc_submission_fini(struct intel_guc *guc) guc_clients_destroy(guc); WARN_ON(!guc_verify_doorbells(guc)); - guc_ads_destroy(guc); - intel_guc_log_destroy(guc); guc_stage_desc_pool_destroy(guc); } diff --git a/drivers/gpu/drm/i915/intel_hangcheck.c b/drivers/gpu/drm/i915/intel_hangcheck.c index 348a4f7ffb67..42e45ae87393 100644 --- a/drivers/gpu/drm/i915/intel_hangcheck.c +++ b/drivers/gpu/drm/i915/intel_hangcheck.c @@ -359,7 +359,7 @@ static void hangcheck_accumulate_sample(struct intel_engine_cs *engine, case ENGINE_DEAD: if (drm_debug & DRM_UT_DRIVER) { struct drm_printer p = drm_debug_printer("hangcheck"); - intel_engine_dump(engine, &p, "%s", engine->name); + intel_engine_dump(engine, &p, "%s\n", engine->name); } break; diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c index 179d0ad3889d..2df18a85c3b4 100644 --- a/drivers/gpu/drm/i915/intel_hdmi.c +++ b/drivers/gpu/drm/i915/intel_hdmi.c @@ -1567,7 +1567,10 @@ intel_hdmi_dp_dual_mode_detect(struct drm_connector *connector, bool has_edid) * there's nothing connected to the port. */ if (type == DRM_DP_DUAL_MODE_UNKNOWN) { - if (has_edid && + /* An overridden EDID imply that we want this port for testing. + * Make sure not to set limits for that port. + */ + if (has_edid && !connector->override_edid && intel_bios_is_port_dp_dual_mode(dev_priv, port)) { DRM_DEBUG_KMS("Assuming DP dual mode adaptor presence based on VBT\n"); type = DRM_DP_DUAL_MODE_TYPE1_DVI; @@ -1932,6 +1935,9 @@ static u8 cnp_port_to_ddc_pin(struct drm_i915_private *dev_priv, case PORT_D: ddc_pin = GMBUS_PIN_4_CNP; break; + case PORT_F: + ddc_pin = GMBUS_PIN_3_BXT; + break; default: MISSING_CASE(port); ddc_pin = GMBUS_PIN_1_BXT; @@ -1940,6 +1946,37 @@ static u8 cnp_port_to_ddc_pin(struct drm_i915_private *dev_priv, return ddc_pin; } +static u8 icl_port_to_ddc_pin(struct drm_i915_private *dev_priv, enum port port) +{ + u8 ddc_pin; + + switch (port) { + case PORT_A: + ddc_pin = GMBUS_PIN_1_BXT; + break; + case PORT_B: + ddc_pin = GMBUS_PIN_2_BXT; + break; + case PORT_C: + ddc_pin = GMBUS_PIN_9_TC1_ICP; + break; + case PORT_D: + ddc_pin = GMBUS_PIN_10_TC2_ICP; + break; + case PORT_E: + ddc_pin = GMBUS_PIN_11_TC3_ICP; + break; + case PORT_F: + ddc_pin = GMBUS_PIN_12_TC4_ICP; + break; + default: + MISSING_CASE(port); + ddc_pin = GMBUS_PIN_2_BXT; + break; + } + return ddc_pin; +} + static u8 g4x_port_to_ddc_pin(struct drm_i915_private *dev_priv, enum port port) { @@ -1982,6 +2019,8 @@ static u8 intel_hdmi_ddc_pin(struct drm_i915_private *dev_priv, ddc_pin = bxt_port_to_ddc_pin(dev_priv, port); else if (HAS_PCH_CNP(dev_priv)) ddc_pin = cnp_port_to_ddc_pin(dev_priv, port); + else if (IS_ICELAKE(dev_priv)) + ddc_pin = icl_port_to_ddc_pin(dev_priv, port); else ddc_pin = g4x_port_to_ddc_pin(dev_priv, port); @@ -2052,7 +2091,7 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port, if (WARN_ON(port == PORT_A)) return; - intel_encoder->hpd_pin = intel_hpd_pin(port); + intel_encoder->hpd_pin = intel_hpd_pin_default(dev_priv, port); if (HAS_DDI(dev_priv)) intel_connector->get_hw_state = intel_ddi_connector_get_hw_state; diff --git a/drivers/gpu/drm/i915/intel_hotplug.c b/drivers/gpu/drm/i915/intel_hotplug.c index 875d5d218d5c..fe28c1ea84a5 100644 --- a/drivers/gpu/drm/i915/intel_hotplug.c +++ b/drivers/gpu/drm/i915/intel_hotplug.c @@ -78,12 +78,14 @@ /** * intel_hpd_port - return port hard associated with certain pin. + * @dev_priv: private driver data pointer * @pin: the hpd pin to get associated port * * Return port that is associatade with @pin and PORT_NONE if no port is * hard associated with that @pin. */ -enum port intel_hpd_pin_to_port(enum hpd_pin pin) +enum port intel_hpd_pin_to_port(struct drm_i915_private *dev_priv, + enum hpd_pin pin) { switch (pin) { case HPD_PORT_A: @@ -95,6 +97,8 @@ enum port intel_hpd_pin_to_port(enum hpd_pin pin) case HPD_PORT_D: return PORT_D; case HPD_PORT_E: + if (IS_CNL_WITH_PORT_F(dev_priv)) + return PORT_F; return PORT_E; default: return PORT_NONE; /* no port for this pin */ @@ -102,13 +106,17 @@ enum port intel_hpd_pin_to_port(enum hpd_pin pin) } /** - * intel_hpd_pin - return pin hard associated with certain port. + * intel_hpd_pin_default - return default pin associated with certain port. + * @dev_priv: private driver data pointer * @port: the hpd port to get associated pin * + * It is only valid and used by digital port encoder. + * * Return pin that is associatade with @port and HDP_NONE if no pin is * hard associated with that @port. */ -enum hpd_pin intel_hpd_pin(enum port port) +enum hpd_pin intel_hpd_pin_default(struct drm_i915_private *dev_priv, + enum port port) { switch (port) { case PORT_A: @@ -121,6 +129,9 @@ enum hpd_pin intel_hpd_pin(enum port port) return HPD_PORT_D; case PORT_E: return HPD_PORT_E; + case PORT_F: + if (IS_CNL_WITH_PORT_F(dev_priv)) + return HPD_PORT_E; default: MISSING_CASE(port); return HPD_NONE; @@ -417,7 +428,7 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv, if (!(BIT(i) & pin_mask)) continue; - port = intel_hpd_pin_to_port(i); + port = intel_hpd_pin_to_port(dev_priv, i); is_dig_port = port != PORT_NONE && dev_priv->hotplug.irq_port[port]; diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c index ef9f91a0b0c9..ad1b1a345f2e 100644 --- a/drivers/gpu/drm/i915/intel_i2c.c +++ b/drivers/gpu/drm/i915/intel_i2c.c @@ -75,11 +75,22 @@ static const struct gmbus_pin gmbus_pins_cnp[] = { [GMBUS_PIN_4_CNP] = { "dpd", GPIOE }, }; +static const struct gmbus_pin gmbus_pins_icp[] = { + [GMBUS_PIN_1_BXT] = { "dpa", GPIOA }, + [GMBUS_PIN_2_BXT] = { "dpb", GPIOB }, + [GMBUS_PIN_9_TC1_ICP] = { "tc1", GPIOC }, + [GMBUS_PIN_10_TC2_ICP] = { "tc2", GPIOD }, + [GMBUS_PIN_11_TC3_ICP] = { "tc3", GPIOE }, + [GMBUS_PIN_12_TC4_ICP] = { "tc4", GPIOF }, +}; + /* pin is expected to be valid */ static const struct gmbus_pin *get_gmbus_pin(struct drm_i915_private *dev_priv, unsigned int pin) { - if (HAS_PCH_CNP(dev_priv)) + if (HAS_PCH_ICP(dev_priv)) + return &gmbus_pins_icp[pin]; + else if (HAS_PCH_CNP(dev_priv)) return &gmbus_pins_cnp[pin]; else if (IS_GEN9_LP(dev_priv)) return &gmbus_pins_bxt[pin]; @@ -96,7 +107,9 @@ bool intel_gmbus_is_valid_pin(struct drm_i915_private *dev_priv, { unsigned int size; - if (HAS_PCH_CNP(dev_priv)) + if (HAS_PCH_ICP(dev_priv)) + size = ARRAY_SIZE(gmbus_pins_icp); + else if (HAS_PCH_CNP(dev_priv)) size = ARRAY_SIZE(gmbus_pins_cnp); else if (IS_GEN9_LP(dev_priv)) size = ARRAY_SIZE(gmbus_pins_bxt); diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index 7ece2f061b9e..380c0838d8b3 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -137,6 +137,7 @@ #include <drm/i915_drm.h> #include "i915_drv.h" #include "i915_gem_render_state.h" +#include "intel_lrc_reg.h" #include "intel_mocs.h" #define RING_EXECLIST_QFULL (1 << 0x2) @@ -156,55 +157,6 @@ #define GEN8_CTX_STATUS_COMPLETED_MASK \ (GEN8_CTX_STATUS_COMPLETE | GEN8_CTX_STATUS_PREEMPTED) -#define CTX_LRI_HEADER_0 0x01 -#define CTX_CONTEXT_CONTROL 0x02 -#define CTX_RING_HEAD 0x04 -#define CTX_RING_TAIL 0x06 -#define CTX_RING_BUFFER_START 0x08 -#define CTX_RING_BUFFER_CONTROL 0x0a -#define CTX_BB_HEAD_U 0x0c -#define CTX_BB_HEAD_L 0x0e -#define CTX_BB_STATE 0x10 -#define CTX_SECOND_BB_HEAD_U 0x12 -#define CTX_SECOND_BB_HEAD_L 0x14 -#define CTX_SECOND_BB_STATE 0x16 -#define CTX_BB_PER_CTX_PTR 0x18 -#define CTX_RCS_INDIRECT_CTX 0x1a -#define CTX_RCS_INDIRECT_CTX_OFFSET 0x1c -#define CTX_LRI_HEADER_1 0x21 -#define CTX_CTX_TIMESTAMP 0x22 -#define CTX_PDP3_UDW 0x24 -#define CTX_PDP3_LDW 0x26 -#define CTX_PDP2_UDW 0x28 -#define CTX_PDP2_LDW 0x2a -#define CTX_PDP1_UDW 0x2c -#define CTX_PDP1_LDW 0x2e -#define CTX_PDP0_UDW 0x30 -#define CTX_PDP0_LDW 0x32 -#define CTX_LRI_HEADER_2 0x41 -#define CTX_R_PWR_CLK_STATE 0x42 -#define CTX_GPGPU_CSR_BASE_ADDRESS 0x44 - -#define CTX_REG(reg_state, pos, reg, val) do { \ - (reg_state)[(pos)+0] = i915_mmio_reg_offset(reg); \ - (reg_state)[(pos)+1] = (val); \ -} while (0) - -#define ASSIGN_CTX_PDP(ppgtt, reg_state, n) do { \ - const u64 _addr = i915_page_dir_dma_addr((ppgtt), (n)); \ - reg_state[CTX_PDP ## n ## _UDW+1] = upper_32_bits(_addr); \ - reg_state[CTX_PDP ## n ## _LDW+1] = lower_32_bits(_addr); \ -} while (0) - -#define ASSIGN_CTX_PML4(ppgtt, reg_state) do { \ - reg_state[CTX_PDP0_UDW + 1] = upper_32_bits(px_dma(&ppgtt->pml4)); \ - reg_state[CTX_PDP0_LDW + 1] = lower_32_bits(px_dma(&ppgtt->pml4)); \ -} while (0) - -#define GEN8_CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT 0x17 -#define GEN9_CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT 0x26 -#define GEN10_CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT 0x19 - /* Typical size of the average request (2 pipecontrols and a MI_BB) */ #define EXECLISTS_REQUEST_SIZE 64 /* bytes */ #define WA_TAIL_DWORDS 2 @@ -504,6 +456,12 @@ static void inject_preempt_context(struct intel_engine_cs *engine) ce->ring->tail &= (ce->ring->size - 1); ce->lrc_reg_state[CTX_RING_TAIL+1] = ce->ring->tail; + GEM_BUG_ON((ce->lrc_reg_state[CTX_CONTEXT_CONTROL + 1] & + _MASKED_BIT_ENABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT | + CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT)) != + _MASKED_BIT_ENABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT | + CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT)); + GEM_TRACE("%s\n", engine->name); for (n = execlists_num_ports(&engine->execlists); --n; ) elsp_write(0, engine->execlists.elsp); @@ -778,6 +736,7 @@ static void execlists_submission_tasklet(unsigned long data) struct intel_engine_execlists * const execlists = &engine->execlists; struct execlist_port * const port = execlists->port; struct drm_i915_private *dev_priv = engine->i915; + bool fw = false; /* We can skip acquiring intel_runtime_pm_get() here as it was taken * on our behalf by the request (see i915_gem_mark_busy()) and it will @@ -788,8 +747,6 @@ static void execlists_submission_tasklet(unsigned long data) */ GEM_BUG_ON(!dev_priv->gt.awake); - intel_uncore_forcewake_get(dev_priv, execlists->fw_domains); - /* Prefer doing test_and_clear_bit() as a two stage operation to avoid * imposing the cost of a locked atomic transaction when submitting a * new request (outside of the context-switch interrupt). @@ -818,6 +775,12 @@ static void execlists_submission_tasklet(unsigned long data) */ __clear_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted); if (unlikely(execlists->csb_head == -1)) { /* following a reset */ + if (!fw) { + intel_uncore_forcewake_get(dev_priv, + execlists->fw_domains); + fw = true; + } + head = readl(dev_priv->regs + i915_mmio_reg_offset(RING_CONTEXT_STATUS_PTR(engine))); tail = GEN8_CSB_WRITE_PTR(head); head = GEN8_CSB_READ_PTR(head); @@ -830,10 +793,10 @@ static void execlists_submission_tasklet(unsigned long data) head = execlists->csb_head; tail = READ_ONCE(buf[write_idx]); } - GEM_TRACE("%s cs-irq head=%d [%d], tail=%d [%d]\n", + GEM_TRACE("%s cs-irq head=%d [%d%s], tail=%d [%d%s]\n", engine->name, - head, GEN8_CSB_READ_PTR(readl(dev_priv->regs + i915_mmio_reg_offset(RING_CONTEXT_STATUS_PTR(engine)))), - tail, GEN8_CSB_WRITE_PTR(readl(dev_priv->regs + i915_mmio_reg_offset(RING_CONTEXT_STATUS_PTR(engine))))); + head, GEN8_CSB_READ_PTR(readl(dev_priv->regs + i915_mmio_reg_offset(RING_CONTEXT_STATUS_PTR(engine)))), fw ? "" : "?", + tail, GEN8_CSB_WRITE_PTR(readl(dev_priv->regs + i915_mmio_reg_offset(RING_CONTEXT_STATUS_PTR(engine)))), fw ? "" : "?"); while (head != tail) { struct drm_i915_gem_request *rq; @@ -943,7 +906,8 @@ static void execlists_submission_tasklet(unsigned long data) if (!execlists_is_active(execlists, EXECLISTS_ACTIVE_PREEMPT)) execlists_dequeue(engine); - intel_uncore_forcewake_put(dev_priv, execlists->fw_domains); + if (fw) + intel_uncore_forcewake_put(dev_priv, execlists->fw_domains); } static void insert_request(struct intel_engine_cs *engine, @@ -1014,7 +978,8 @@ static void execlists_schedule(struct drm_i915_gem_request *request, int prio) stack.signaler = &request->priotree; list_add(&stack.dfs_link, &dfs); - /* Recursively bump all dependent priorities to match the new request. + /* + * Recursively bump all dependent priorities to match the new request. * * A naive approach would be to use recursion: * static void update_priorities(struct i915_priotree *pt, prio) { @@ -1031,27 +996,29 @@ static void execlists_schedule(struct drm_i915_gem_request *request, int prio) * end result is a topological list of requests in reverse order, the * last element in the list is the request we must execute first. */ - list_for_each_entry_safe(dep, p, &dfs, dfs_link) { + list_for_each_entry(dep, &dfs, dfs_link) { struct i915_priotree *pt = dep->signaler; - /* Within an engine, there can be no cycle, but we may + /* + * Within an engine, there can be no cycle, but we may * refer to the same dependency chain multiple times * (redundant dependencies are not eliminated) and across * engines. */ list_for_each_entry(p, &pt->signalers_list, signal_link) { - if (i915_gem_request_completed(pt_to_request(p->signaler))) + GEM_BUG_ON(p == dep); /* no cycles! */ + + if (i915_priotree_signaled(p->signaler)) continue; GEM_BUG_ON(p->signaler->priority < pt->priority); if (prio > READ_ONCE(p->signaler->priority)) list_move_tail(&p->dfs_link, &dfs); } - - list_safe_reset_next(dep, p, dfs_link); } - /* If we didn't need to bump any existing priorities, and we haven't + /* + * If we didn't need to bump any existing priorities, and we haven't * yet submitted this request (i.e. there is no potential race with * execlists_submit_request()), we can set our own priority and skip * acquiring the engine locks. @@ -1125,11 +1092,9 @@ execlists_context_pin(struct intel_engine_cs *engine, goto out; GEM_BUG_ON(!ce->pin_count); /* no overflow please! */ - if (!ce->state) { - ret = execlists_context_deferred_alloc(ctx, engine); - if (ret) - goto err; - } + ret = execlists_context_deferred_alloc(ctx, engine); + if (ret) + goto err; GEM_BUG_ON(!ce->state); ret = __context_pin(ctx, ce->state); @@ -1363,6 +1328,40 @@ static u32 *gen9_init_indirectctx_bb(struct intel_engine_cs *engine, u32 *batch) return batch; } +static u32 * +gen10_init_indirectctx_bb(struct intel_engine_cs *engine, u32 *batch) +{ + int i; + + /* + * WaPipeControlBefore3DStateSamplePattern: cnl + * + * Ensure the engine is idle prior to programming a + * 3DSTATE_SAMPLE_PATTERN during a context restore. + */ + batch = gen8_emit_pipe_control(batch, + PIPE_CONTROL_CS_STALL, + 0); + /* + * WaPipeControlBefore3DStateSamplePattern says we need 4 dwords for + * the PIPE_CONTROL followed by 12 dwords of 0x0, so 16 dwords in + * total. However, a PIPE_CONTROL is 6 dwords long, not 4, which is + * confusing. Since gen8_emit_pipe_control() already advances the + * batch by 6 dwords, we advance the other 10 here, completing a + * cacheline. It's not clear if the workaround requires this padding + * before other commands, or if it's just the regular padding we would + * already have for the workaround bb, so leave it here for now. + */ + for (i = 0; i < 10; i++) + *batch++ = MI_NOOP; + + /* Pad to end of cacheline */ + while ((unsigned long)batch % CACHELINE_BYTES) + *batch++ = MI_NOOP; + + return batch; +} + #define CTX_WA_BB_OBJ_SIZE (PAGE_SIZE) static int lrc_setup_wa_ctx(struct intel_engine_cs *engine) @@ -1411,12 +1410,14 @@ static int intel_init_workaround_bb(struct intel_engine_cs *engine) unsigned int i; int ret; - if (WARN_ON(engine->id != RCS || !engine->scratch)) + if (GEM_WARN_ON(engine->id != RCS)) return -EINVAL; switch (INTEL_GEN(engine->i915)) { case 10: - return 0; + wa_bb_fn[0] = gen10_init_indirectctx_bb; + wa_bb_fn[1] = NULL; + break; case 9: wa_bb_fn[0] = gen9_init_indirectctx_bb; wa_bb_fn[1] = NULL; @@ -1446,7 +1447,8 @@ static int intel_init_workaround_bb(struct intel_engine_cs *engine) */ for (i = 0; i < ARRAY_SIZE(wa_bb_fn); i++) { wa_bb[i]->offset = batch_ptr - batch; - if (WARN_ON(!IS_ALIGNED(wa_bb[i]->offset, CACHELINE_BYTES))) { + if (GEM_WARN_ON(!IS_ALIGNED(wa_bb[i]->offset, + CACHELINE_BYTES))) { ret = -EINVAL; break; } @@ -1472,47 +1474,48 @@ static u8 gtiir[] = { [VECS] = 3, }; -static int gen8_init_common_ring(struct intel_engine_cs *engine) +static void enable_execlists(struct intel_engine_cs *engine) { struct drm_i915_private *dev_priv = engine->i915; - struct intel_engine_execlists * const execlists = &engine->execlists; - int ret; - ret = intel_mocs_init_engine(engine); - if (ret) - return ret; + I915_WRITE(RING_HWSTAM(engine->mmio_base), 0xffffffff); - intel_engine_reset_breadcrumbs(engine); - intel_engine_init_hangcheck(engine); + /* + * Make sure we're not enabling the new 12-deep CSB + * FIFO as that requires a slightly updated handling + * in the ctx switch irq. Since we're currently only + * using only 2 elements of the enhanced execlists the + * deeper FIFO it's not needed and it's not worth adding + * more statements to the irq handler to support it. + */ + if (INTEL_GEN(dev_priv) >= 11) + I915_WRITE(RING_MODE_GEN7(engine), + _MASKED_BIT_DISABLE(GEN11_GFX_DISABLE_LEGACY_MODE)); + else + I915_WRITE(RING_MODE_GEN7(engine), + _MASKED_BIT_ENABLE(GFX_RUN_LIST_ENABLE)); - I915_WRITE(RING_HWSTAM(engine->mmio_base), 0xffffffff); - I915_WRITE(RING_MODE_GEN7(engine), - _MASKED_BIT_ENABLE(GFX_RUN_LIST_ENABLE)); I915_WRITE(RING_HWS_PGA(engine->mmio_base), engine->status_page.ggtt_offset); POSTING_READ(RING_HWS_PGA(engine->mmio_base)); - DRM_DEBUG_DRIVER("Execlists enabled for %s\n", engine->name); + /* Following the reset, we need to reload the CSB read/write pointers */ + engine->execlists.csb_head = -1; +} - GEM_BUG_ON(engine->id >= ARRAY_SIZE(gtiir)); +static int gen8_init_common_ring(struct intel_engine_cs *engine) +{ + struct intel_engine_execlists * const execlists = &engine->execlists; + int ret; - /* - * Clear any pending interrupt state. - * - * We do it twice out of paranoia that some of the IIR are double - * buffered, and if we only reset it once there may still be - * an interrupt pending. - */ - I915_WRITE(GEN8_GT_IIR(gtiir[engine->id]), - GT_CONTEXT_SWITCH_INTERRUPT << engine->irq_shift); - I915_WRITE(GEN8_GT_IIR(gtiir[engine->id]), - GT_CONTEXT_SWITCH_INTERRUPT << engine->irq_shift); - clear_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted); - execlists->csb_head = -1; - execlists->active = 0; + ret = intel_mocs_init_engine(engine); + if (ret) + return ret; + + intel_engine_reset_breadcrumbs(engine); + intel_engine_init_hangcheck(engine); - execlists->elsp = - dev_priv->regs + i915_mmio_reg_offset(RING_ELSP(engine)); + enable_execlists(engine); /* After a GPU reset, we may have requests to replay */ if (execlists->first) @@ -1554,6 +1557,31 @@ static int gen9_init_render_ring(struct intel_engine_cs *engine) return init_workarounds_ring(engine); } +static void reset_irq(struct intel_engine_cs *engine) +{ + struct drm_i915_private *dev_priv = engine->i915; + int i; + + GEM_BUG_ON(engine->id >= ARRAY_SIZE(gtiir)); + + /* + * Clear any pending interrupt state. + * + * We do it twice out of paranoia that some of the IIR are double + * buffered, and if we only reset it once there may still be + * an interrupt pending. + */ + for (i = 0; i < 2; i++) { + I915_WRITE(GEN8_GT_IIR(gtiir[engine->id]), + GT_CONTEXT_SWITCH_INTERRUPT << engine->irq_shift); + POSTING_READ(GEN8_GT_IIR(gtiir[engine->id])); + } + GEM_BUG_ON(I915_READ(GEN8_GT_IIR(gtiir[engine->id])) & + (GT_CONTEXT_SWITCH_INTERRUPT << engine->irq_shift)); + + clear_bit(ENGINE_IRQ_EXECLIST, &engine->irq_posted); +} + static void reset_common_ring(struct intel_engine_cs *engine, struct drm_i915_gem_request *request) { @@ -1563,6 +1591,9 @@ static void reset_common_ring(struct intel_engine_cs *engine, GEM_TRACE("%s seqno=%x\n", engine->name, request ? request->global_seqno : 0); + + reset_irq(engine); + spin_lock_irqsave(&engine->timeline->lock, flags); /* @@ -1581,6 +1612,9 @@ static void reset_common_ring(struct intel_engine_cs *engine, spin_unlock_irqrestore(&engine->timeline->lock, flags); + /* Mark all CS interrupts as complete */ + execlists->active = 0; + /* If the request was innocent, we leave the request in the ELSP * and will try to replay it on restarting. The context image may * have been corrupted by the reset, in which case we may have @@ -1912,6 +1946,7 @@ void intel_logical_ring_cleanup(struct intel_engine_cs *engine) intel_engine_cleanup_common(engine); lrc_destroy_wa_ctx(engine); + engine->i915 = NULL; dev_priv->engine[engine->id] = NULL; kfree(engine); @@ -2001,6 +2036,9 @@ static int logical_ring_init(struct intel_engine_cs *engine) if (ret) goto error; + engine->execlists.elsp = + engine->i915->regs + i915_mmio_reg_offset(RING_ELSP(engine)); + return 0; error: @@ -2142,6 +2180,8 @@ static void execlists_init_reg_state(u32 *regs, MI_LRI_FORCE_POSTED; CTX_REG(regs, CTX_CONTEXT_CONTROL, RING_CONTEXT_CONTROL(engine), + _MASKED_BIT_DISABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT | + CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT) | _MASKED_BIT_ENABLE(CTX_CTRL_INHIBIT_SYN_CTX_SWITCH | (HAS_RESOURCE_STREAMER(dev_priv) ? CTX_CTRL_RS_CTX_ENABLE : 0))); @@ -2261,6 +2301,10 @@ populate_lr_context(struct i915_gem_context *ctx, if (!engine->default_state) regs[CTX_CONTEXT_CONTROL + 1] |= _MASKED_BIT_ENABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT); + if (ctx->hw_id == PREEMPT_ID) + regs[CTX_CONTEXT_CONTROL + 1] |= + _MASKED_BIT_ENABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT | + CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT); i915_gem_object_unpin_map(ctx_obj); @@ -2277,7 +2321,8 @@ static int execlists_context_deferred_alloc(struct i915_gem_context *ctx, struct intel_ring *ring; int ret; - WARN_ON(ce->state); + if (ce->state) + return 0; context_size = round_up(engine->context_size, I915_GTT_PAGE_SIZE); diff --git a/drivers/gpu/drm/i915/intel_lrc.h b/drivers/gpu/drm/i915/intel_lrc.h index 6d4f9b995a11..636ced41225d 100644 --- a/drivers/gpu/drm/i915/intel_lrc.h +++ b/drivers/gpu/drm/i915/intel_lrc.h @@ -37,6 +37,7 @@ #define CTX_CTRL_INHIBIT_SYN_CTX_SWITCH (1 << 3) #define CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT (1 << 0) #define CTX_CTRL_RS_CTX_ENABLE (1 << 1) +#define CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT (1 << 2) #define RING_CONTEXT_STATUS_BUF_BASE(engine) _MMIO((engine)->mmio_base + 0x370) #define RING_CONTEXT_STATUS_BUF_LO(engine, i) _MMIO((engine)->mmio_base + 0x370 + (i) * 8) #define RING_CONTEXT_STATUS_BUF_HI(engine, i) _MMIO((engine)->mmio_base + 0x370 + (i) * 8 + 4) diff --git a/drivers/gpu/drm/i915/intel_lrc_reg.h b/drivers/gpu/drm/i915/intel_lrc_reg.h new file mode 100644 index 000000000000..a53336e2fc97 --- /dev/null +++ b/drivers/gpu/drm/i915/intel_lrc_reg.h @@ -0,0 +1,67 @@ +/* + * SPDX-License-Identifier: MIT + * + * Copyright © 2014-2018 Intel Corporation + */ + +#ifndef _INTEL_LRC_REG_H_ +#define _INTEL_LRC_REG_H_ + +#include <linux/types.h> + +/* GEN8+ Reg State Context */ +#define CTX_LRI_HEADER_0 0x01 +#define CTX_CONTEXT_CONTROL 0x02 +#define CTX_RING_HEAD 0x04 +#define CTX_RING_TAIL 0x06 +#define CTX_RING_BUFFER_START 0x08 +#define CTX_RING_BUFFER_CONTROL 0x0a +#define CTX_BB_HEAD_U 0x0c +#define CTX_BB_HEAD_L 0x0e +#define CTX_BB_STATE 0x10 +#define CTX_SECOND_BB_HEAD_U 0x12 +#define CTX_SECOND_BB_HEAD_L 0x14 +#define CTX_SECOND_BB_STATE 0x16 +#define CTX_BB_PER_CTX_PTR 0x18 +#define CTX_RCS_INDIRECT_CTX 0x1a +#define CTX_RCS_INDIRECT_CTX_OFFSET 0x1c +#define CTX_LRI_HEADER_1 0x21 +#define CTX_CTX_TIMESTAMP 0x22 +#define CTX_PDP3_UDW 0x24 +#define CTX_PDP3_LDW 0x26 +#define CTX_PDP2_UDW 0x28 +#define CTX_PDP2_LDW 0x2a +#define CTX_PDP1_UDW 0x2c +#define CTX_PDP1_LDW 0x2e +#define CTX_PDP0_UDW 0x30 +#define CTX_PDP0_LDW 0x32 +#define CTX_LRI_HEADER_2 0x41 +#define CTX_R_PWR_CLK_STATE 0x42 +#define CTX_GPGPU_CSR_BASE_ADDRESS 0x44 + +#define CTX_REG(reg_state, pos, reg, val) do { \ + u32 *reg_state__ = (reg_state); \ + const u32 pos__ = (pos); \ + (reg_state__)[(pos__) + 0] = i915_mmio_reg_offset(reg); \ + (reg_state__)[(pos__) + 1] = (val); \ +} while (0) + +#define ASSIGN_CTX_PDP(ppgtt, reg_state, n) do { \ + u32 *reg_state__ = (reg_state); \ + const u64 addr__ = i915_page_dir_dma_addr((ppgtt), (n)); \ + (reg_state__)[CTX_PDP ## n ## _UDW + 1] = upper_32_bits(addr__); \ + (reg_state__)[CTX_PDP ## n ## _LDW + 1] = lower_32_bits(addr__); \ +} while (0) + +#define ASSIGN_CTX_PML4(ppgtt, reg_state) do { \ + u32 *reg_state__ = (reg_state); \ + const u64 addr__ = px_dma(&ppgtt->pml4); \ + (reg_state__)[CTX_PDP0_UDW + 1] = upper_32_bits(addr__); \ + (reg_state__)[CTX_PDP0_LDW + 1] = lower_32_bits(addr__); \ +} while (0) + +#define GEN8_CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT 0x17 +#define GEN9_CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT 0x26 +#define GEN10_CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT 0x19 + +#endif /* _INTEL_LRC_REG_H_ */ diff --git a/drivers/gpu/drm/i915/intel_lspcon.c b/drivers/gpu/drm/i915/intel_lspcon.c index dcbc786479f9..8ae8f42f430a 100644 --- a/drivers/gpu/drm/i915/intel_lspcon.c +++ b/drivers/gpu/drm/i915/intel_lspcon.c @@ -167,11 +167,10 @@ static void lspcon_resume_in_pcon_wa(struct intel_lspcon *lspcon) { struct intel_dp *intel_dp = lspcon_to_intel_dp(lspcon); struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp); - struct drm_i915_private *dev_priv = to_i915(dig_port->base.base.dev); unsigned long start = jiffies; while (1) { - if (intel_digital_port_connected(dev_priv, dig_port)) { + if (intel_digital_port_connected(&dig_port->base)) { DRM_DEBUG_KMS("LSPCON recovering in PCON mode after %u ms\n", jiffies_to_msecs(jiffies - start)); return; diff --git a/drivers/gpu/drm/i915/intel_modes.c b/drivers/gpu/drm/i915/intel_modes.c index 4e43f873c889..b39846613e3c 100644 --- a/drivers/gpu/drm/i915/intel_modes.c +++ b/drivers/gpu/drm/i915/intel_modes.c @@ -30,21 +30,6 @@ #include "intel_drv.h" #include "i915_drv.h" -static void intel_connector_update_eld_conn_type(struct drm_connector *connector) -{ - u8 conn_type; - - if (connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort || - connector->connector_type == DRM_MODE_CONNECTOR_eDP) { - conn_type = DRM_ELD_CONN_TYPE_DP; - } else { - conn_type = DRM_ELD_CONN_TYPE_HDMI; - } - - connector->eld[DRM_ELD_SAD_COUNT_CONN_TYPE] &= ~DRM_ELD_CONN_TYPE_MASK; - connector->eld[DRM_ELD_SAD_COUNT_CONN_TYPE] |= conn_type; -} - /** * intel_connector_update_modes - update connector from edid * @connector: DRM connector device to use @@ -58,8 +43,6 @@ int intel_connector_update_modes(struct drm_connector *connector, drm_mode_connector_update_edid_property(connector, edid); ret = drm_add_edid_modes(connector, edid); - intel_connector_update_eld_conn_type(connector); - return ret; } diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c index fa6831f8c004..e702a6487aa9 100644 --- a/drivers/gpu/drm/i915/intel_panel.c +++ b/drivers/gpu/drm/i915/intel_panel.c @@ -1719,9 +1719,9 @@ cnp_setup_backlight(struct intel_connector *connector, enum pipe unused) u32 pwm_ctl, val; /* - * CNP has the BXT implementation of backlight, but with only - * one controller. Future platforms could have multiple controllers - * so let's make this extensible and prepared for the future. + * CNP has the BXT implementation of backlight, but with only one + * controller. TODO: ICP has multiple controllers but we only use + * controller 0 for now. */ panel->backlight.controller = 0; @@ -1865,7 +1865,7 @@ intel_panel_init_backlight_funcs(struct intel_panel *panel) panel->backlight.set = bxt_set_backlight; panel->backlight.get = bxt_get_backlight; panel->backlight.hz_to_pwm = bxt_hz_to_pwm; - } else if (HAS_PCH_CNP(dev_priv)) { + } else if (HAS_PCH_CNP(dev_priv) || HAS_PCH_ICP(dev_priv)) { panel->backlight.setup = cnp_setup_backlight; panel->backlight.enable = cnp_enable_backlight; panel->backlight.disable = cnp_disable_backlight; diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 1a6e699e19e0..eb68abf6a8e9 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -3694,11 +3694,18 @@ bool intel_can_enable_sagv(struct drm_atomic_state *state) struct intel_crtc_state *cstate; enum pipe pipe; int level, latency; - int sagv_block_time_us = IS_GEN9(dev_priv) ? 30 : 20; + int sagv_block_time_us; if (!intel_has_sagv(dev_priv)) return false; + if (IS_GEN9(dev_priv)) + sagv_block_time_us = 30; + else if (IS_GEN10(dev_priv)) + sagv_block_time_us = 20; + else + sagv_block_time_us = 10; + /* * SKL+ workaround: bspec recommends we disable the SAGV when we have * more then one pipe enabled @@ -3778,7 +3785,8 @@ skl_ddb_get_pipe_allocation_limits(struct drm_device *dev, ddb_size = INTEL_INFO(dev_priv)->ddb_size; WARN_ON(ddb_size == 0); - ddb_size -= 4; /* 4 blocks for bypass path allocation */ + if (INTEL_GEN(dev_priv) < 11) + ddb_size -= 4; /* 4 blocks for bypass path allocation */ /* * If the state doesn't change the active CRTC's, then there's @@ -4311,7 +4319,7 @@ skl_allocate_pipe_ddb(struct intel_crtc_state *cstate, */ static uint_fixed_16_16_t skl_wm_method1(const struct drm_i915_private *dev_priv, uint32_t pixel_rate, - uint8_t cpp, uint32_t latency) + uint8_t cpp, uint32_t latency, uint32_t dbuf_block_size) { uint32_t wm_intermediate_val; uint_fixed_16_16_t ret; @@ -4320,7 +4328,7 @@ skl_wm_method1(const struct drm_i915_private *dev_priv, uint32_t pixel_rate, return FP_16_16_MAX; wm_intermediate_val = latency * pixel_rate * cpp; - ret = div_fixed16(wm_intermediate_val, 1000 * 512); + ret = div_fixed16(wm_intermediate_val, 1000 * dbuf_block_size); if (INTEL_GEN(dev_priv) >= 10) ret = add_fixed16_u32(ret, 1); @@ -4430,6 +4438,12 @@ skl_compute_plane_wm_params(const struct drm_i915_private *dev_priv, wp->plane_pixel_rate = skl_adjusted_plane_pixel_rate(cstate, intel_pstate); + if (INTEL_GEN(dev_priv) >= 11 && + fb->modifier == I915_FORMAT_MOD_Yf_TILED && wp->cpp == 8) + wp->dbuf_block_size = 256; + else + wp->dbuf_block_size = 512; + if (drm_rotation_90_or_270(pstate->rotation)) { switch (wp->cpp) { @@ -4456,7 +4470,8 @@ skl_compute_plane_wm_params(const struct drm_i915_private *dev_priv, wp->plane_bytes_per_line = wp->width * wp->cpp; if (wp->y_tiled) { interm_pbpl = DIV_ROUND_UP(wp->plane_bytes_per_line * - wp->y_min_scanlines, 512); + wp->y_min_scanlines, + wp->dbuf_block_size); if (INTEL_GEN(dev_priv) >= 10) interm_pbpl++; @@ -4464,10 +4479,12 @@ skl_compute_plane_wm_params(const struct drm_i915_private *dev_priv, wp->plane_blocks_per_line = div_fixed16(interm_pbpl, wp->y_min_scanlines); } else if (wp->x_tiled && IS_GEN9(dev_priv)) { - interm_pbpl = DIV_ROUND_UP(wp->plane_bytes_per_line, 512); + interm_pbpl = DIV_ROUND_UP(wp->plane_bytes_per_line, + wp->dbuf_block_size); wp->plane_blocks_per_line = u32_to_fixed16(interm_pbpl); } else { - interm_pbpl = DIV_ROUND_UP(wp->plane_bytes_per_line, 512) + 1; + interm_pbpl = DIV_ROUND_UP(wp->plane_bytes_per_line, + wp->dbuf_block_size) + 1; wp->plane_blocks_per_line = u32_to_fixed16(interm_pbpl); } @@ -4497,6 +4514,7 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, struct intel_atomic_state *state = to_intel_atomic_state(cstate->base.state); bool apply_memory_bw_wa = skl_needs_memory_bw_wa(state); + uint32_t min_disp_buf_needed; if (latency == 0 || !intel_wm_plane_visible(cstate, intel_pstate)) { @@ -4514,7 +4532,7 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, latency += 15; method1 = skl_wm_method1(dev_priv, wp->plane_pixel_rate, - wp->cpp, latency); + wp->cpp, latency, wp->dbuf_block_size); method2 = skl_wm_method2(wp->plane_pixel_rate, cstate->base.adjusted_mode.crtc_htotal, latency, @@ -4524,7 +4542,8 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, selected_result = max_fixed16(method2, wp->y_tile_minimum); } else { if ((wp->cpp * cstate->base.adjusted_mode.crtc_htotal / - 512 < 1) && (wp->plane_bytes_per_line / 512 < 1)) + wp->dbuf_block_size < 1) && + (wp->plane_bytes_per_line / wp->dbuf_block_size < 1)) selected_result = method2; else if (ddb_allocation >= fixed16_to_u32_round_up(wp->plane_blocks_per_line)) @@ -4554,7 +4573,31 @@ static int skl_compute_plane_wm(const struct drm_i915_private *dev_priv, } } - if (res_blocks >= ddb_allocation || res_lines > 31) { + if (INTEL_GEN(dev_priv) >= 11) { + if (wp->y_tiled) { + uint32_t extra_lines; + uint_fixed_16_16_t fp_min_disp_buf_needed; + + if (res_lines % wp->y_min_scanlines == 0) + extra_lines = wp->y_min_scanlines; + else + extra_lines = wp->y_min_scanlines * 2 - + res_lines % wp->y_min_scanlines; + + fp_min_disp_buf_needed = mul_u32_fixed16(res_lines + + extra_lines, + wp->plane_blocks_per_line); + min_disp_buf_needed = fixed16_to_u32_round_up( + fp_min_disp_buf_needed); + } else { + min_disp_buf_needed = DIV_ROUND_UP(res_blocks * 11, 10); + } + } else { + min_disp_buf_needed = res_blocks; + } + + if (res_blocks >= ddb_allocation || res_lines > 31 || + min_disp_buf_needed >= ddb_allocation) { *enabled = false; /* @@ -4790,8 +4833,10 @@ static void skl_write_plane_wm(struct intel_crtc *intel_crtc, skl_ddb_entry_write(dev_priv, PLANE_BUF_CFG(pipe, plane_id), &ddb->plane[pipe][plane_id]); - skl_ddb_entry_write(dev_priv, PLANE_NV12_BUF_CFG(pipe, plane_id), - &ddb->y_plane[pipe][plane_id]); + if (INTEL_GEN(dev_priv) < 11) + skl_ddb_entry_write(dev_priv, + PLANE_NV12_BUF_CFG(pipe, plane_id), + &ddb->y_plane[pipe][plane_id]); } static void skl_write_cursor_wm(struct intel_crtc *intel_crtc, @@ -6626,9 +6671,29 @@ static void gen9_enable_rc6(struct drm_i915_private *dev_priv) I915_WRITE(GEN6_RC_SLEEP, 0); - /* 2c: Program Coarse Power Gating Policies. */ - I915_WRITE(GEN9_MEDIA_PG_IDLE_HYSTERESIS, 25); - I915_WRITE(GEN9_RENDER_PG_IDLE_HYSTERESIS, 25); + /* + * 2c: Program Coarse Power Gating Policies. + * + * Bspec's guidance is to use 25us (really 25 * 1280ns) here. What we + * use instead is a more conservative estimate for the maximum time + * it takes us to service a CS interrupt and submit a new ELSP - that + * is the time which the GPU is idle waiting for the CPU to select the + * next request to execute. If the idle hysteresis is less than that + * interrupt service latency, the hardware will automatically gate + * the power well and we will then incur the wake up cost on top of + * the service latency. A similar guide from intel_pstate is that we + * do not want the enable hysteresis to less than the wakeup latency. + * + * igt/gem_exec_nop/sequential provides a rough estimate for the + * service latency, and puts it around 10us for Broadwell (and other + * big core) and around 40us for Broxton (and other low power cores). + * [Note that for legacy ringbuffer submission, this is less than 1us!] + * However, the wakeup latency on Broxton is closer to 100us. To be + * conservative, we have to factor in a context switch on top (due + * to ksoftirqd). + */ + I915_WRITE(GEN9_MEDIA_PG_IDLE_HYSTERESIS, 250); + I915_WRITE(GEN9_RENDER_PG_IDLE_HYSTERESIS, 250); /* 3a: Enable RC6 */ I915_WRITE(GEN6_RC6_THRESHOLD, 37500); /* 37.5/125ms per EI */ @@ -9150,7 +9215,8 @@ int sandybridge_pcode_read(struct drm_i915_private *dev_priv, u32 mbox, u32 *val } int sandybridge_pcode_write_timeout(struct drm_i915_private *dev_priv, - u32 mbox, u32 val, int timeout_us) + u32 mbox, u32 val, + int fast_timeout_us, int slow_timeout_ms) { int status; @@ -9173,7 +9239,8 @@ int sandybridge_pcode_write_timeout(struct drm_i915_private *dev_priv, if (__intel_wait_for_register_fw(dev_priv, GEN6_PCODE_MAILBOX, GEN6_PCODE_READY, 0, - timeout_us, 0, NULL)) { + fast_timeout_us, slow_timeout_ms, + NULL)) { DRM_ERROR("timeout waiting for pcode write of 0x%08x to mbox %x to finish for %ps\n", val, mbox, __builtin_return_address(0)); return -ETIMEDOUT; diff --git a/drivers/gpu/drm/i915/intel_psr.c b/drivers/gpu/drm/i915/intel_psr.c index 2e32615eeada..e9feffdea899 100644 --- a/drivers/gpu/drm/i915/intel_psr.c +++ b/drivers/gpu/drm/i915/intel_psr.c @@ -56,14 +56,6 @@ #include "intel_drv.h" #include "i915_drv.h" -static bool is_edp_psr(struct intel_dp *intel_dp) -{ - if (!intel_dp_is_edp(intel_dp)) - return false; - - return intel_dp->psr_dpcd[0] & DP_PSR_IS_SUPPORTED; -} - static bool vlv_is_psr_active_on_pipe(struct drm_device *dev, int pipe) { struct drm_i915_private *dev_priv = to_i915(dev); @@ -358,10 +350,7 @@ void intel_psr_compute_config(struct intel_dp *intel_dp, &crtc_state->base.adjusted_mode; int psr_setup_time; - if (!HAS_PSR(dev_priv)) - return; - - if (!is_edp_psr(intel_dp)) + if (!CAN_PSR(dev_priv)) return; if (!i915_modparams.enable_psr) { @@ -476,7 +465,7 @@ static void hsw_psr_enable_source(struct intel_dp *intel_dp, chicken |= PSR2_ADD_VERTICAL_LINE_COUNT; I915_WRITE(CHICKEN_TRANS(cpu_transcoder), chicken); - I915_WRITE(EDP_PSR_DEBUG_CTL, + I915_WRITE(EDP_PSR_DEBUG, EDP_PSR_DEBUG_MASK_MEMUP | EDP_PSR_DEBUG_MASK_HPD | EDP_PSR_DEBUG_MASK_LPSP | @@ -490,7 +479,7 @@ static void hsw_psr_enable_source(struct intel_dp *intel_dp, * preventing other hw tracking issues now we can rely * on frontbuffer tracking. */ - I915_WRITE(EDP_PSR_DEBUG_CTL, + I915_WRITE(EDP_PSR_DEBUG, EDP_PSR_DEBUG_MASK_MEMUP | EDP_PSR_DEBUG_MASK_HPD | EDP_PSR_DEBUG_MASK_LPSP); @@ -514,6 +503,9 @@ void intel_psr_enable(struct intel_dp *intel_dp, if (!crtc_state->has_psr) return; + if (WARN_ON(!CAN_PSR(dev_priv))) + return; + WARN_ON(dev_priv->drrs.dp); mutex_lock(&dev_priv->psr.lock); if (dev_priv->psr.enabled) { @@ -522,8 +514,6 @@ void intel_psr_enable(struct intel_dp *intel_dp, } dev_priv->psr.psr2_support = crtc_state->has_psr2; - dev_priv->psr.source_ok = true; - dev_priv->psr.busy_frontbuffer_bits = 0; dev_priv->psr.setup_vsc(intel_dp, crtc_state); @@ -599,7 +589,7 @@ static void hsw_psr_disable(struct intel_dp *intel_dp, 0); if (dev_priv->psr.psr2_support) { - psr_status = EDP_PSR2_STATUS_CTL; + psr_status = EDP_PSR2_STATUS; psr_status_mask = EDP_PSR2_STATUS_STATE_MASK; I915_WRITE(EDP_PSR2_CTL, @@ -607,7 +597,7 @@ static void hsw_psr_disable(struct intel_dp *intel_dp, ~(EDP_PSR2_ENABLE | EDP_SU_TRACK_ENABLE)); } else { - psr_status = EDP_PSR_STATUS_CTL; + psr_status = EDP_PSR_STATUS; psr_status_mask = EDP_PSR_STATUS_STATE_MASK; I915_WRITE(EDP_PSR_CTL, @@ -646,6 +636,9 @@ void intel_psr_disable(struct intel_dp *intel_dp, if (!old_crtc_state->has_psr) return; + if (WARN_ON(!CAN_PSR(dev_priv))) + return; + mutex_lock(&dev_priv->psr.lock); if (!dev_priv->psr.enabled) { mutex_unlock(&dev_priv->psr.lock); @@ -679,19 +672,19 @@ static void intel_psr_work(struct work_struct *work) if (HAS_DDI(dev_priv)) { if (dev_priv->psr.psr2_support) { if (intel_wait_for_register(dev_priv, - EDP_PSR2_STATUS_CTL, - EDP_PSR2_STATUS_STATE_MASK, - 0, - 50)) { + EDP_PSR2_STATUS, + EDP_PSR2_STATUS_STATE_MASK, + 0, + 50)) { DRM_ERROR("Timed out waiting for PSR2 Idle for re-enable\n"); return; } } else { if (intel_wait_for_register(dev_priv, - EDP_PSR_STATUS_CTL, - EDP_PSR_STATUS_STATE_MASK, - 0, - 50)) { + EDP_PSR_STATUS, + EDP_PSR_STATUS_STATE_MASK, + 0, + 50)) { DRM_ERROR("Timed out waiting for PSR Idle for re-enable\n"); return; } @@ -796,7 +789,7 @@ void intel_psr_single_frame_update(struct drm_i915_private *dev_priv, enum pipe pipe; u32 val; - if (!HAS_PSR(dev_priv)) + if (!CAN_PSR(dev_priv)) return; /* @@ -845,7 +838,7 @@ void intel_psr_invalidate(struct drm_i915_private *dev_priv, struct drm_crtc *crtc; enum pipe pipe; - if (!HAS_PSR(dev_priv)) + if (!CAN_PSR(dev_priv)) return; mutex_lock(&dev_priv->psr.lock); @@ -885,7 +878,7 @@ void intel_psr_flush(struct drm_i915_private *dev_priv, struct drm_crtc *crtc; enum pipe pipe; - if (!HAS_PSR(dev_priv)) + if (!CAN_PSR(dev_priv)) return; mutex_lock(&dev_priv->psr.lock); @@ -926,6 +919,9 @@ void intel_psr_init(struct drm_i915_private *dev_priv) dev_priv->psr_mmio_base = IS_HASWELL(dev_priv) ? HSW_EDP_PSR_BASE : BDW_EDP_PSR_BASE; + if (!dev_priv->psr.sink_support) + return; + /* Per platform default: all disabled. */ if (i915_modparams.enable_psr == -1) i915_modparams.enable_psr = 0; diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index c5ff203e42d6..a0e7a6c2a57c 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -366,20 +366,6 @@ struct intel_engine_cs { */ #define I915_ENGINE_SAMPLE_MAX (I915_SAMPLE_SEMA + 1) struct i915_pmu_sample sample[I915_ENGINE_SAMPLE_MAX]; - /** - * @busy_stats: Has enablement of engine stats tracking been - * requested. - */ - bool busy_stats; - /** - * @disable_busy_stats: Work item for busy stats disabling. - * - * Same as with @enable_busy_stats action, with the difference - * that we delay it in case there are rapid enable-disable - * actions, which can happen during tool startup (like perf - * stat). - */ - struct delayed_work disable_busy_stats; } pmu; /* diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c index d758da6156a8..70e659772a7a 100644 --- a/drivers/gpu/drm/i915/intel_runtime_pm.c +++ b/drivers/gpu/drm/i915/intel_runtime_pm.c @@ -94,6 +94,8 @@ intel_display_power_domain_str(enum intel_display_power_domain domain) return "PORT_DDI_D_LANES"; case POWER_DOMAIN_PORT_DDI_E_LANES: return "PORT_DDI_E_LANES"; + case POWER_DOMAIN_PORT_DDI_F_LANES: + return "PORT_DDI_F_LANES"; case POWER_DOMAIN_PORT_DDI_A_IO: return "PORT_DDI_A_IO"; case POWER_DOMAIN_PORT_DDI_B_IO: @@ -104,6 +106,8 @@ intel_display_power_domain_str(enum intel_display_power_domain domain) return "PORT_DDI_D_IO"; case POWER_DOMAIN_PORT_DDI_E_IO: return "PORT_DDI_E_IO"; + case POWER_DOMAIN_PORT_DDI_F_IO: + return "PORT_DDI_F_IO"; case POWER_DOMAIN_PORT_DSI: return "PORT_DSI"; case POWER_DOMAIN_PORT_CRT: @@ -124,6 +128,8 @@ intel_display_power_domain_str(enum intel_display_power_domain domain) return "AUX_C"; case POWER_DOMAIN_AUX_D: return "AUX_D"; + case POWER_DOMAIN_AUX_F: + return "AUX_F"; case POWER_DOMAIN_GMBUS: return "GMBUS"; case POWER_DOMAIN_INIT: @@ -390,6 +396,15 @@ static void hsw_power_well_enable(struct drm_i915_private *dev_priv, I915_WRITE(HSW_PWR_WELL_CTL_DRIVER(id), val | HSW_PWR_WELL_CTL_REQ(id)); hsw_wait_for_power_well_enable(dev_priv, power_well); + /* Display WA #1178: cnl */ + if (IS_CANNONLAKE(dev_priv) && + (id == CNL_DISP_PW_AUX_B || id == CNL_DISP_PW_AUX_C || + id == CNL_DISP_PW_AUX_D || id == CNL_DISP_PW_AUX_F)) { + val = I915_READ(CNL_AUX_ANAOVRD1(id)); + val |= CNL_AUX_ANAOVRD1_ENABLE | CNL_AUX_ANAOVRD1_LDO_BYPASS; + I915_WRITE(CNL_AUX_ANAOVRD1(id), val); + } + if (wait_fuses) gen9_wait_for_power_well_fuses(dev_priv, pg); @@ -1816,9 +1831,11 @@ void intel_display_power_put(struct drm_i915_private *dev_priv, BIT_ULL(POWER_DOMAIN_PORT_DDI_B_LANES) | \ BIT_ULL(POWER_DOMAIN_PORT_DDI_C_LANES) | \ BIT_ULL(POWER_DOMAIN_PORT_DDI_D_LANES) | \ + BIT_ULL(POWER_DOMAIN_PORT_DDI_F_LANES) | \ BIT_ULL(POWER_DOMAIN_AUX_B) | \ BIT_ULL(POWER_DOMAIN_AUX_C) | \ BIT_ULL(POWER_DOMAIN_AUX_D) | \ + BIT_ULL(POWER_DOMAIN_AUX_F) | \ BIT_ULL(POWER_DOMAIN_AUDIO) | \ BIT_ULL(POWER_DOMAIN_VGA) | \ BIT_ULL(POWER_DOMAIN_INIT)) @@ -1846,8 +1863,15 @@ void intel_display_power_put(struct drm_i915_private *dev_priv, #define CNL_DISPLAY_AUX_D_POWER_DOMAINS ( \ BIT_ULL(POWER_DOMAIN_AUX_D) | \ BIT_ULL(POWER_DOMAIN_INIT)) +#define CNL_DISPLAY_AUX_F_POWER_DOMAINS ( \ + BIT_ULL(POWER_DOMAIN_AUX_F) | \ + BIT_ULL(POWER_DOMAIN_INIT)) +#define CNL_DISPLAY_DDI_F_IO_POWER_DOMAINS ( \ + BIT_ULL(POWER_DOMAIN_PORT_DDI_F_IO) | \ + BIT_ULL(POWER_DOMAIN_INIT)) #define CNL_DISPLAY_DC_OFF_POWER_DOMAINS ( \ CNL_DISPLAY_POWERWELL_2_POWER_DOMAINS | \ + BIT_ULL(POWER_DOMAIN_GT_IRQ) | \ BIT_ULL(POWER_DOMAIN_MODESET) | \ BIT_ULL(POWER_DOMAIN_AUX_A) | \ BIT_ULL(POWER_DOMAIN_INIT)) @@ -2395,6 +2419,18 @@ static struct i915_power_well cnl_power_wells[] = { .ops = &hsw_power_well_ops, .id = SKL_DISP_PW_DDI_D, }, + { + .name = "DDI F IO power well", + .domains = CNL_DISPLAY_DDI_F_IO_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = CNL_DISP_PW_DDI_F, + }, + { + .name = "AUX F", + .domains = CNL_DISPLAY_AUX_F_POWER_DOMAINS, + .ops = &hsw_power_well_ops, + .id = CNL_DISP_PW_AUX_F, + }, }; static int @@ -2510,6 +2546,16 @@ int intel_power_domains_init(struct drm_i915_private *dev_priv) set_power_wells(power_domains, skl_power_wells); } else if (IS_CANNONLAKE(dev_priv)) { set_power_wells(power_domains, cnl_power_wells); + + /* + * DDI and Aux IO are getting enabled for all ports + * regardless the presence or use. So, in order to avoid + * timeouts, lets remove them from the list + * for the SKUs without port F. + */ + if (!IS_CNL_WITH_PORT_F(dev_priv)) + power_domains->power_well_count -= 2; + } else if (IS_BROXTON(dev_priv)) { set_power_wells(power_domains, bxt_power_wells); } else if (IS_GEMINILAKE(dev_priv)) { diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c index dd485f59eb1d..32f10621fac8 100644 --- a/drivers/gpu/drm/i915/intel_sprite.c +++ b/drivers/gpu/drm/i915/intel_sprite.c @@ -865,6 +865,7 @@ intel_check_sprite_plane(struct intel_plane *plane, struct drm_rect *src = &state->base.src; struct drm_rect *dst = &state->base.dst; const struct drm_rect *clip = &state->clip; + int max_stride = INTEL_GEN(dev_priv) >= 9 ? 32768 : 16384; int hscale, vscale; int max_scale, min_scale; bool can_scale; @@ -885,7 +886,7 @@ intel_check_sprite_plane(struct intel_plane *plane, } /* FIXME check all gen limits */ - if (fb->width < 3 || fb->height < 3 || fb->pitches[0] > 16384) { + if (fb->width < 3 || fb->height < 3 || fb->pitches[0] > max_stride) { DRM_DEBUG_KMS("Unsuitable framebuffer for plane\n"); return -EINVAL; } @@ -893,7 +894,7 @@ intel_check_sprite_plane(struct intel_plane *plane, /* setup can_scale, min_scale, max_scale */ if (INTEL_GEN(dev_priv) >= 9) { /* use scaler when colorkey is not required */ - if (state->ckey.flags == I915_SET_COLORKEY_NONE) { + if (!state->ckey.flags) { can_scale = 1; min_scale = 1; max_scale = skl_max_scale(crtc, crtc_state); @@ -1027,7 +1028,7 @@ intel_check_sprite_plane(struct intel_plane *plane, dst->y2 = crtc_y + crtc_h; if (INTEL_GEN(dev_priv) >= 9) { - ret = skl_check_plane_surface(state); + ret = skl_check_plane_surface(crtc_state, state); if (ret) return ret; @@ -1069,6 +1070,9 @@ int intel_sprite_set_colorkey(struct drm_device *dev, void *data, struct drm_modeset_acquire_ctx ctx; int ret = 0; + /* ignore the pointless "none" flag */ + set->flags &= ~I915_SET_COLORKEY_NONE; + /* Make sure we don't try to enable both src & dest simultaneously */ if ((set->flags & (I915_SET_COLORKEY_DESTINATION | I915_SET_COLORKEY_SOURCE)) == (I915_SET_COLORKEY_DESTINATION | I915_SET_COLORKEY_SOURCE)) return -EINVAL; @@ -1161,18 +1165,27 @@ static uint32_t skl_plane_formats[] = { DRM_FORMAT_VYUY, }; -static const uint64_t skl_plane_format_modifiers[] = { +static const uint64_t skl_plane_format_modifiers_noccs[] = { + I915_FORMAT_MOD_Yf_TILED, + I915_FORMAT_MOD_Y_TILED, I915_FORMAT_MOD_X_TILED, DRM_FORMAT_MOD_LINEAR, DRM_FORMAT_MOD_INVALID }; -static bool g4x_sprite_plane_format_mod_supported(struct drm_plane *plane, - uint32_t format, - uint64_t modifier) +static const uint64_t skl_plane_format_modifiers_ccs[] = { + I915_FORMAT_MOD_Yf_TILED_CCS, + I915_FORMAT_MOD_Y_TILED_CCS, + I915_FORMAT_MOD_Yf_TILED, + I915_FORMAT_MOD_Y_TILED, + I915_FORMAT_MOD_X_TILED, + DRM_FORMAT_MOD_LINEAR, + DRM_FORMAT_MOD_INVALID +}; + +static bool g4x_mod_supported(uint32_t format, uint64_t modifier) { switch (format) { - case DRM_FORMAT_XBGR8888: case DRM_FORMAT_XRGB8888: case DRM_FORMAT_YUYV: case DRM_FORMAT_YVYU: @@ -1187,22 +1200,38 @@ static bool g4x_sprite_plane_format_mod_supported(struct drm_plane *plane, } } -static bool vlv_sprite_plane_format_mod_supported(struct drm_plane *plane, - uint32_t format, - uint64_t modifier) +static bool snb_mod_supported(uint32_t format, uint64_t modifier) { switch (format) { + case DRM_FORMAT_XRGB8888: + case DRM_FORMAT_XBGR8888: case DRM_FORMAT_YUYV: case DRM_FORMAT_YVYU: case DRM_FORMAT_UYVY: case DRM_FORMAT_VYUY: + if (modifier == DRM_FORMAT_MOD_LINEAR || + modifier == I915_FORMAT_MOD_X_TILED) + return true; + /* fall through */ + default: + return false; + } +} + +static bool vlv_mod_supported(uint32_t format, uint64_t modifier) +{ + switch (format) { case DRM_FORMAT_RGB565: - case DRM_FORMAT_XRGB8888: + case DRM_FORMAT_ABGR8888: case DRM_FORMAT_ARGB8888: + case DRM_FORMAT_XBGR8888: + case DRM_FORMAT_XRGB8888: case DRM_FORMAT_XBGR2101010: case DRM_FORMAT_ABGR2101010: - case DRM_FORMAT_XBGR8888: - case DRM_FORMAT_ABGR8888: + case DRM_FORMAT_YUYV: + case DRM_FORMAT_YVYU: + case DRM_FORMAT_UYVY: + case DRM_FORMAT_VYUY: if (modifier == DRM_FORMAT_MOD_LINEAR || modifier == I915_FORMAT_MOD_X_TILED) return true; @@ -1212,16 +1241,17 @@ static bool vlv_sprite_plane_format_mod_supported(struct drm_plane *plane, } } -static bool skl_sprite_plane_format_mod_supported(struct drm_plane *plane, - uint32_t format, - uint64_t modifier) +static bool skl_mod_supported(uint32_t format, uint64_t modifier) { - /* This is the same as primary plane since SKL has universal planes */ switch (format) { case DRM_FORMAT_XRGB8888: case DRM_FORMAT_XBGR8888: case DRM_FORMAT_ARGB8888: case DRM_FORMAT_ABGR8888: + if (modifier == I915_FORMAT_MOD_Yf_TILED_CCS || + modifier == I915_FORMAT_MOD_Y_TILED_CCS) + return true; + /* fall through */ case DRM_FORMAT_RGB565: case DRM_FORMAT_XRGB2101010: case DRM_FORMAT_XBGR2101010: @@ -1257,13 +1287,13 @@ static bool intel_sprite_plane_format_mod_supported(struct drm_plane *plane, return false; if (INTEL_GEN(dev_priv) >= 9) - return skl_sprite_plane_format_mod_supported(plane, format, modifier); + return skl_mod_supported(format, modifier); else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) - return vlv_sprite_plane_format_mod_supported(plane, format, modifier); + return vlv_mod_supported(format, modifier); + else if (INTEL_GEN(dev_priv) >= 6) + return snb_mod_supported(format, modifier); else - return g4x_sprite_plane_format_mod_supported(plane, format, modifier); - - unreachable(); + return g4x_mod_supported(format, modifier); } static const struct drm_plane_funcs intel_sprite_plane_funcs = { @@ -1277,6 +1307,23 @@ static const struct drm_plane_funcs intel_sprite_plane_funcs = { .format_mod_supported = intel_sprite_plane_format_mod_supported, }; +bool skl_plane_has_ccs(struct drm_i915_private *dev_priv, + enum pipe pipe, enum plane_id plane_id) +{ + if (plane_id == PLANE_CURSOR) + return false; + + if (INTEL_GEN(dev_priv) >= 10) + return true; + + if (IS_GEMINILAKE(dev_priv)) + return pipe != PIPE_C; + + return pipe != PIPE_C && + (plane_id == PLANE_PRIMARY || + plane_id == PLANE_SPRITE0); +} + struct intel_plane * intel_sprite_plane_create(struct drm_i915_private *dev_priv, enum pipe pipe, int plane) @@ -1303,7 +1350,7 @@ intel_sprite_plane_create(struct drm_i915_private *dev_priv, } intel_plane->base.state = &state->base; - if (INTEL_GEN(dev_priv) >= 10) { + if (INTEL_GEN(dev_priv) >= 9) { intel_plane->can_scale = true; state->scaler_id = -1; @@ -1313,18 +1360,11 @@ intel_sprite_plane_create(struct drm_i915_private *dev_priv, plane_formats = skl_plane_formats; num_plane_formats = ARRAY_SIZE(skl_plane_formats); - modifiers = skl_plane_format_modifiers; - } else if (INTEL_GEN(dev_priv) >= 9) { - intel_plane->can_scale = true; - state->scaler_id = -1; - intel_plane->update_plane = skl_update_plane; - intel_plane->disable_plane = skl_disable_plane; - intel_plane->get_hw_state = skl_plane_get_hw_state; - - plane_formats = skl_plane_formats; - num_plane_formats = ARRAY_SIZE(skl_plane_formats); - modifiers = skl_plane_format_modifiers; + if (skl_plane_has_ccs(dev_priv, pipe, PLANE_SPRITE0 + plane)) + modifiers = skl_plane_format_modifiers_ccs; + else + modifiers = skl_plane_format_modifiers_noccs; } else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv)) { intel_plane->can_scale = false; intel_plane->max_downscale = 1; @@ -1386,7 +1426,7 @@ intel_sprite_plane_create(struct drm_i915_private *dev_priv, intel_plane->pipe = pipe; intel_plane->i9xx_plane = plane; intel_plane->id = PLANE_SPRITE0 + plane; - intel_plane->frontbuffer_bit = INTEL_FRONTBUFFER_SPRITE(pipe, plane); + intel_plane->frontbuffer_bit = INTEL_FRONTBUFFER(pipe, intel_plane->id); intel_plane->check_plane = intel_check_sprite_plane; possible_crtcs = (1 << pipe); diff --git a/drivers/gpu/drm/i915/intel_uc.c b/drivers/gpu/drm/i915/intel_uc.c index d82ca0f438f5..9f1bac6398fb 100644 --- a/drivers/gpu/drm/i915/intel_uc.c +++ b/drivers/gpu/drm/i915/intel_uc.c @@ -27,6 +27,8 @@ #include "intel_guc.h" #include "i915_drv.h" +static void guc_free_load_err_log(struct intel_guc *guc); + /* Reset GuC providing us with fresh state for both GuC and HuC. */ static int __intel_uc_reset_hw(struct drm_i915_private *dev_priv) @@ -65,6 +67,21 @@ static int __get_platform_enable_guc(struct drm_i915_private *dev_priv) return enable_guc; } +static int __get_default_guc_log_level(struct drm_i915_private *dev_priv) +{ + int guc_log_level = 0; /* disabled */ + + /* Enable if we're running on platform with GuC and debug config */ + if (HAS_GUC(dev_priv) && intel_uc_is_using_guc() && + (IS_ENABLED(CONFIG_DRM_I915_DEBUG) || + IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM))) + guc_log_level = 1 + GUC_LOG_VERBOSITY_MAX; + + /* Any platform specific fine-tuning can be done here */ + + return guc_log_level; +} + /** * intel_uc_sanitize_options - sanitize uC related modparam options * @dev_priv: device private @@ -74,6 +91,13 @@ static int __get_platform_enable_guc(struct drm_i915_private *dev_priv) * modparam varies between platforms and it is hardcoded in driver code. * Any other modparam value is only monitored against availability of the * related hardware or firmware definitions. + * + * In case of "guc_log_level" option this function will attempt to modify + * it only if it was initially set to "auto(-1)" or if initial value was + * "enable(1..4)" on platforms without the GuC. Default value for this + * modparam varies between platforms and is usually set to "disable(0)" + * unless GuC is enabled on given platform and the driver is compiled with + * debug config when this modparam will default to "enable(1..4)". */ void intel_uc_sanitize_options(struct drm_i915_private *dev_priv) { @@ -91,22 +115,48 @@ void intel_uc_sanitize_options(struct drm_i915_private *dev_priv) /* Verify GuC firmware availability */ if (intel_uc_is_using_guc() && !intel_uc_fw_is_selected(guc_fw)) { - DRM_WARN("Incompatible option detected: enable_guc=%d, %s!\n", - i915_modparams.enable_guc, + DRM_WARN("Incompatible option detected: %s=%d, %s!\n", + "enable_guc", i915_modparams.enable_guc, !HAS_GUC(dev_priv) ? "no GuC hardware" : "no GuC firmware"); } /* Verify HuC firmware availability */ if (intel_uc_is_using_huc() && !intel_uc_fw_is_selected(huc_fw)) { - DRM_WARN("Incompatible option detected: enable_guc=%d, %s!\n", - i915_modparams.enable_guc, + DRM_WARN("Incompatible option detected: %s=%d, %s!\n", + "enable_guc", i915_modparams.enable_guc, !HAS_HUC(dev_priv) ? "no HuC hardware" : "no HuC firmware"); } + /* A negative value means "use platform/config default" */ + if (i915_modparams.guc_log_level < 0) + i915_modparams.guc_log_level = + __get_default_guc_log_level(dev_priv); + + if (i915_modparams.guc_log_level > 0 && !intel_uc_is_using_guc()) { + DRM_WARN("Incompatible option detected: %s=%d, %s!\n", + "guc_log_level", i915_modparams.guc_log_level, + !HAS_GUC(dev_priv) ? "no GuC hardware" : + "GuC not enabled"); + i915_modparams.guc_log_level = 0; + } + + if (i915_modparams.guc_log_level > 1 + GUC_LOG_VERBOSITY_MAX) { + DRM_WARN("Incompatible option detected: %s=%d, %s!\n", + "guc_log_level", i915_modparams.guc_log_level, + "verbosity too high"); + i915_modparams.guc_log_level = 1 + GUC_LOG_VERBOSITY_MAX; + } + + DRM_DEBUG_DRIVER("guc_log_level=%d (enabled:%s verbosity:%d)\n", + i915_modparams.guc_log_level, + yesno(i915_modparams.guc_log_level), + i915_modparams.guc_log_level - 1); + /* Make sure that sanitization was done */ GEM_BUG_ON(i915_modparams.enable_guc < 0); + GEM_BUG_ON(i915_modparams.guc_log_level < 0); } void intel_uc_init_early(struct drm_i915_private *dev_priv) @@ -135,6 +185,8 @@ void intel_uc_fini_fw(struct drm_i915_private *dev_priv) if (USES_HUC(dev_priv)) intel_uc_fw_fini(&dev_priv->huc.fw); + + guc_free_load_err_log(&dev_priv->guc); } /** @@ -152,7 +204,7 @@ void intel_uc_init_mmio(struct drm_i915_private *dev_priv) static void guc_capture_load_err_log(struct intel_guc *guc) { - if (!guc->log.vma || i915_modparams.guc_log_level < 0) + if (!guc->log.vma || !i915_modparams.guc_log_level) return; if (!guc->load_err_log) @@ -188,28 +240,44 @@ static void guc_disable_communication(struct intel_guc *guc) guc->send = intel_guc_send_nop; } -int intel_uc_init_wq(struct drm_i915_private *dev_priv) +int intel_uc_init_misc(struct drm_i915_private *dev_priv) { + struct intel_guc *guc = &dev_priv->guc; int ret; if (!USES_GUC(dev_priv)) return 0; - ret = intel_guc_init_wq(&dev_priv->guc); + ret = intel_guc_init_wq(guc); if (ret) { DRM_ERROR("Couldn't allocate workqueues for GuC\n"); - return ret; + goto err; + } + + ret = intel_guc_log_relay_create(guc); + if (ret) { + DRM_ERROR("Couldn't allocate relay for GuC log\n"); + goto err_relay; } return 0; + +err_relay: + intel_guc_fini_wq(guc); +err: + return ret; } -void intel_uc_fini_wq(struct drm_i915_private *dev_priv) +void intel_uc_fini_misc(struct drm_i915_private *dev_priv) { + struct intel_guc *guc = &dev_priv->guc; + if (!USES_GUC(dev_priv)) return; - intel_guc_fini_wq(&dev_priv->guc); + intel_guc_fini_wq(guc); + + intel_guc_log_relay_destroy(guc); } int intel_uc_init(struct drm_i915_private *dev_priv) @@ -322,7 +390,7 @@ int intel_uc_init_hw(struct drm_i915_private *dev_priv) } if (USES_GUC_SUBMISSION(dev_priv)) { - if (i915_modparams.guc_log_level >= 0) + if (i915_modparams.guc_log_level) gen9_enable_guc_interrupts(dev_priv); ret = intel_guc_submission_enable(guc); @@ -364,8 +432,6 @@ void intel_uc_fini_hw(struct drm_i915_private *dev_priv) { struct intel_guc *guc = &dev_priv->guc; - guc_free_load_err_log(guc); - if (!USES_GUC(dev_priv)) return; diff --git a/drivers/gpu/drm/i915/intel_uc.h b/drivers/gpu/drm/i915/intel_uc.h index 8a7249722ef1..f2984e01e257 100644 --- a/drivers/gpu/drm/i915/intel_uc.h +++ b/drivers/gpu/drm/i915/intel_uc.h @@ -33,8 +33,8 @@ void intel_uc_init_early(struct drm_i915_private *dev_priv); void intel_uc_init_mmio(struct drm_i915_private *dev_priv); void intel_uc_init_fw(struct drm_i915_private *dev_priv); void intel_uc_fini_fw(struct drm_i915_private *dev_priv); -int intel_uc_init_wq(struct drm_i915_private *dev_priv); -void intel_uc_fini_wq(struct drm_i915_private *dev_priv); +int intel_uc_init_misc(struct drm_i915_private *dev_priv); +void intel_uc_fini_misc(struct drm_i915_private *dev_priv); int intel_uc_init_hw(struct drm_i915_private *dev_priv); void intel_uc_fini_hw(struct drm_i915_private *dev_priv); int intel_uc_init(struct drm_i915_private *dev_priv); diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index 89547b614aa6..4075010ac088 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -1452,7 +1452,7 @@ static const struct reg_whitelist { } reg_read_whitelist[] = { { .offset_ldw = RING_TIMESTAMP(RENDER_RING_BASE), .offset_udw = RING_TIMESTAMP_UDW(RENDER_RING_BASE), - .gen_mask = INTEL_GEN_MASK(4, 10), + .gen_mask = INTEL_GEN_MASK(4, 11), .size = 8 } }; @@ -1936,8 +1936,7 @@ int intel_reset_guc(struct drm_i915_private *dev_priv) { int ret; - if (!HAS_GUC(dev_priv)) - return -EINVAL; + GEM_BUG_ON(!HAS_GUC(dev_priv)); intel_uncore_forcewake_get(dev_priv, FORCEWAKE_ALL); ret = gen6_hw_domain_reset(dev_priv, GEN9_GRDOM_GUC); diff --git a/drivers/gpu/drm/i915/intel_vbt_defs.h b/drivers/gpu/drm/i915/intel_vbt_defs.h index 98dff6058d3c..458468237b5f 100644 --- a/drivers/gpu/drm/i915/intel_vbt_defs.h +++ b/drivers/gpu/drm/i915/intel_vbt_defs.h @@ -227,7 +227,7 @@ struct bdb_general_features { #define DEVICE_TYPE_COMPOSITE_OUTPUT (1 << 9) #define DEVICE_TYPE_DUAL_CHANNEL (1 << 8) #define DEVICE_TYPE_HIGH_SPEED_LINK (1 << 6) -#define DEVICE_TYPE_LVDS_SINGALING (1 << 5) +#define DEVICE_TYPE_LVDS_SIGNALING (1 << 5) #define DEVICE_TYPE_TMDS_DVI_SIGNALING (1 << 4) #define DEVICE_TYPE_VIDEO_SIGNALING (1 << 3) #define DEVICE_TYPE_DISPLAYPORT_OUTPUT (1 << 2) @@ -243,7 +243,7 @@ struct bdb_general_features { DEVICE_TYPE_MIPI_OUTPUT | \ DEVICE_TYPE_COMPOSITE_OUTPUT | \ DEVICE_TYPE_DUAL_CHANNEL | \ - DEVICE_TYPE_LVDS_SINGALING | \ + DEVICE_TYPE_LVDS_SIGNALING | \ DEVICE_TYPE_TMDS_DVI_SIGNALING | \ DEVICE_TYPE_VIDEO_SIGNALING | \ DEVICE_TYPE_DISPLAYPORT_OUTPUT | \ @@ -253,7 +253,7 @@ struct bdb_general_features { (DEVICE_TYPE_INTERNAL_CONNECTOR | \ DEVICE_TYPE_MIPI_OUTPUT | \ DEVICE_TYPE_COMPOSITE_OUTPUT | \ - DEVICE_TYPE_LVDS_SINGALING | \ + DEVICE_TYPE_LVDS_SIGNALING | \ DEVICE_TYPE_TMDS_DVI_SIGNALING | \ DEVICE_TYPE_VIDEO_SIGNALING | \ DEVICE_TYPE_DISPLAYPORT_OUTPUT | \ @@ -299,6 +299,8 @@ struct bdb_general_features { #define DVO_PORT_DPA 10 #define DVO_PORT_DPE 11 /* 193 */ #define DVO_PORT_HDMIE 12 /* 193 */ +#define DVO_PORT_DPF 13 /* N/A */ +#define DVO_PORT_HDMIF 14 /* N/A */ #define DVO_PORT_MIPIA 21 /* 171 */ #define DVO_PORT_MIPIB 22 /* 171 */ #define DVO_PORT_MIPIC 23 /* 171 */ @@ -318,6 +320,11 @@ enum vbt_gmbus_ddi { DDC_BUS_DDI_F, }; +#define VBT_DP_MAX_LINK_RATE_HBR3 0 +#define VBT_DP_MAX_LINK_RATE_HBR2 1 +#define VBT_DP_MAX_LINK_RATE_HBR 2 +#define VBT_DP_MAX_LINK_RATE_LBR 3 + /* * The child device config, aka the display device data structure, provides a * description of a port and its configuration on the platform. diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c index 4a28d713a7d8..d8064276431c 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c @@ -885,6 +885,84 @@ static int shrink_hole(struct drm_i915_private *i915, return err; } +static int shrink_boom(struct drm_i915_private *i915, + struct i915_address_space *vm, + u64 hole_start, u64 hole_end, + unsigned long end_time) +{ + unsigned int sizes[] = { SZ_2M, SZ_1G }; + struct drm_i915_gem_object *purge; + struct drm_i915_gem_object *explode; + int err; + int i; + + /* + * Catch the case which shrink_hole seems to miss. The setup here + * requires invoking the shrinker as we do the alloc_pt/alloc_pd, while + * ensuring that all vma assiocated with the respective pd/pdp are + * unpinned at the time. + */ + + for (i = 0; i < ARRAY_SIZE(sizes); ++i) { + unsigned int flags = PIN_USER | PIN_OFFSET_FIXED; + unsigned int size = sizes[i]; + struct i915_vma *vma; + + purge = fake_dma_object(i915, size); + if (IS_ERR(purge)) + return PTR_ERR(purge); + + vma = i915_vma_instance(purge, vm, NULL); + if (IS_ERR(vma)) { + err = PTR_ERR(vma); + goto err_purge; + } + + err = i915_vma_pin(vma, 0, 0, flags); + if (err) + goto err_purge; + + /* Should now be ripe for purging */ + i915_vma_unpin(vma); + + explode = fake_dma_object(i915, size); + if (IS_ERR(explode)) { + err = PTR_ERR(purge); + goto err_purge; + } + + vm->fault_attr.probability = 100; + vm->fault_attr.interval = 1; + atomic_set(&vm->fault_attr.times, -1); + + vma = i915_vma_instance(explode, vm, NULL); + if (IS_ERR(vma)) { + err = PTR_ERR(vma); + goto err_explode; + } + + err = i915_vma_pin(vma, 0, 0, flags | size); + if (err) + goto err_explode; + + i915_vma_unpin(vma); + + i915_gem_object_put(purge); + i915_gem_object_put(explode); + + memset(&vm->fault_attr, 0, sizeof(vm->fault_attr)); + } + + return 0; + +err_explode: + i915_gem_object_put(explode); +err_purge: + i915_gem_object_put(purge); + memset(&vm->fault_attr, 0, sizeof(vm->fault_attr)); + return err; +} + static int exercise_ppgtt(struct drm_i915_private *dev_priv, int (*func)(struct drm_i915_private *i915, struct i915_address_space *vm, @@ -953,6 +1031,11 @@ static int igt_ppgtt_shrink(void *arg) return exercise_ppgtt(arg, shrink_hole); } +static int igt_ppgtt_shrink_boom(void *arg) +{ + return exercise_ppgtt(arg, shrink_boom); +} + static int sort_holes(void *priv, struct list_head *A, struct list_head *B) { struct drm_mm_node *a = list_entry(A, typeof(*a), hole_stack); @@ -1052,35 +1135,38 @@ static int igt_ggtt_page(void *arg) memset(&tmp, 0, sizeof(tmp)); err = drm_mm_insert_node_in_range(&ggtt->base.mm, &tmp, - 1024 * PAGE_SIZE, 0, + count * PAGE_SIZE, 0, I915_COLOR_UNEVICTABLE, 0, ggtt->mappable_end, DRM_MM_INSERT_LOW); if (err) goto out_unpin; + intel_runtime_pm_get(i915); + + for (n = 0; n < count; n++) { + u64 offset = tmp.start + n * PAGE_SIZE; + + ggtt->base.insert_page(&ggtt->base, + i915_gem_object_get_dma_address(obj, 0), + offset, I915_CACHE_NONE, 0); + } + order = i915_random_order(count, &prng); if (!order) { err = -ENOMEM; goto out_remove; } - intel_runtime_pm_get(i915); for (n = 0; n < count; n++) { u64 offset = tmp.start + order[n] * PAGE_SIZE; u32 __iomem *vaddr; - ggtt->base.insert_page(&ggtt->base, - i915_gem_object_get_dma_address(obj, 0), - offset, I915_CACHE_NONE, 0); - vaddr = io_mapping_map_atomic_wc(&ggtt->iomap, offset); iowrite32(n, vaddr + n); io_mapping_unmap_atomic(vaddr); - - wmb(); - ggtt->base.clear_range(&ggtt->base, offset, PAGE_SIZE); } + i915_gem_flush_ggtt_writes(i915); i915_random_reorder(order, count, &prng); for (n = 0; n < count; n++) { @@ -1088,16 +1174,10 @@ static int igt_ggtt_page(void *arg) u32 __iomem *vaddr; u32 val; - ggtt->base.insert_page(&ggtt->base, - i915_gem_object_get_dma_address(obj, 0), - offset, I915_CACHE_NONE, 0); - vaddr = io_mapping_map_atomic_wc(&ggtt->iomap, offset); val = ioread32(vaddr + n); io_mapping_unmap_atomic(vaddr); - ggtt->base.clear_range(&ggtt->base, offset, PAGE_SIZE); - if (val != n) { pr_err("insert page failed: found %d, expected %d\n", val, n); @@ -1105,10 +1185,11 @@ static int igt_ggtt_page(void *arg) break; } } - intel_runtime_pm_put(i915); kfree(order); out_remove: + ggtt->base.clear_range(&ggtt->base, tmp.start, tmp.size); + intel_runtime_pm_put(i915); drm_mm_remove_node(&tmp); out_unpin: i915_gem_object_unpin_pages(obj); @@ -1579,6 +1660,7 @@ int i915_gem_gtt_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_ppgtt_pot), SUBTEST(igt_ppgtt_fill), SUBTEST(igt_ppgtt_shrink), + SUBTEST(igt_ppgtt_shrink_boom), SUBTEST(igt_ggtt_lowlevel), SUBTEST(igt_ggtt_drunk), SUBTEST(igt_ggtt_walk), diff --git a/drivers/gpu/drm/i915/selftests/i915_random.c b/drivers/gpu/drm/i915/selftests/i915_random.c index 2088ae57aa89..1f415ce47018 100644 --- a/drivers/gpu/drm/i915/selftests/i915_random.c +++ b/drivers/gpu/drm/i915/selftests/i915_random.c @@ -57,7 +57,8 @@ unsigned int *i915_random_order(unsigned int count, struct rnd_state *state) { unsigned int *order, i; - order = kmalloc_array(count, sizeof(*order), GFP_KERNEL | __GFP_NOWARN); + order = kmalloc_array(count, sizeof(*order), + GFP_KERNEL | __GFP_RETRY_MAYFAIL | __GFP_NOWARN); if (!order) return order; diff --git a/drivers/gpu/drm/i915/selftests/i915_sw_fence.c b/drivers/gpu/drm/i915/selftests/i915_sw_fence.c index ea01d0fe3ace..570e325af93e 100644 --- a/drivers/gpu/drm/i915/selftests/i915_sw_fence.c +++ b/drivers/gpu/drm/i915/selftests/i915_sw_fence.c @@ -606,6 +606,139 @@ err: return -EINVAL; } +static const char *mock_name(struct dma_fence *fence) +{ + return "mock"; +} + +static bool mock_enable_signaling(struct dma_fence *fence) +{ + return true; +} + +static const struct dma_fence_ops mock_fence_ops = { + .get_driver_name = mock_name, + .get_timeline_name = mock_name, + .enable_signaling = mock_enable_signaling, + .wait = dma_fence_default_wait, + .release = dma_fence_free, +}; + +static DEFINE_SPINLOCK(mock_fence_lock); + +static struct dma_fence *alloc_dma_fence(void) +{ + struct dma_fence *dma; + + dma = kmalloc(sizeof(*dma), GFP_KERNEL); + if (dma) + dma_fence_init(dma, &mock_fence_ops, &mock_fence_lock, 0, 0); + + return dma; +} + +static struct i915_sw_fence * +wrap_dma_fence(struct dma_fence *dma, unsigned long delay) +{ + struct i915_sw_fence *fence; + int err; + + fence = alloc_fence(); + if (!fence) + return ERR_PTR(-ENOMEM); + + err = i915_sw_fence_await_dma_fence(fence, dma, delay, GFP_NOWAIT); + i915_sw_fence_commit(fence); + if (err < 0) { + free_fence(fence); + return ERR_PTR(err); + } + + return fence; +} + +static int test_dma_fence(void *arg) +{ + struct i915_sw_fence *timeout = NULL, *not = NULL; + unsigned long delay = i915_selftest.timeout_jiffies; + unsigned long end, sleep; + struct dma_fence *dma; + int err; + + dma = alloc_dma_fence(); + if (!dma) + return -ENOMEM; + + timeout = wrap_dma_fence(dma, delay); + if (IS_ERR(timeout)) { + err = PTR_ERR(timeout); + goto err; + } + + not = wrap_dma_fence(dma, 0); + if (IS_ERR(not)) { + err = PTR_ERR(not); + goto err; + } + + err = -EINVAL; + if (i915_sw_fence_done(timeout) || i915_sw_fence_done(not)) { + pr_err("Fences immediately signaled\n"); + goto err; + } + + /* We round the timeout for the fence up to the next second */ + end = round_jiffies_up(jiffies + delay); + + sleep = jiffies_to_usecs(delay) / 3; + usleep_range(sleep, 2 * sleep); + if (time_after(jiffies, end)) { + pr_debug("Slept too long, delay=%lu, (target=%lu, now=%lu) skipping\n", + delay, end, jiffies); + goto skip; + } + + if (i915_sw_fence_done(timeout) || i915_sw_fence_done(not)) { + pr_err("Fences signaled too early\n"); + goto err; + } + + if (!wait_event_timeout(timeout->wait, + i915_sw_fence_done(timeout), + 2 * (end - jiffies) + 1)) { + pr_err("Timeout fence unsignaled!\n"); + goto err; + } + + if (i915_sw_fence_done(not)) { + pr_err("No timeout fence signaled!\n"); + goto err; + } + +skip: + dma_fence_signal(dma); + + if (!i915_sw_fence_done(timeout) || !i915_sw_fence_done(not)) { + pr_err("Fences unsignaled\n"); + goto err; + } + + free_fence(not); + free_fence(timeout); + dma_fence_put(dma); + + return 0; + +err: + dma_fence_signal(dma); + if (!IS_ERR_OR_NULL(timeout)) + free_fence(timeout); + if (!IS_ERR_OR_NULL(not)) + free_fence(not); + dma_fence_put(dma); + return err; +} + int i915_sw_fence_mock_selftests(void) { static const struct i915_subtest tests[] = { @@ -618,6 +751,7 @@ int i915_sw_fence_mock_selftests(void) SUBTEST(test_chain), SUBTEST(test_ipc), SUBTEST(test_timer), + SUBTEST(test_dma_fence), }; return i915_subtests(tests, NULL); diff --git a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c index d1f91a533afa..d1d2c2456f69 100644 --- a/drivers/gpu/drm/i915/selftests/intel_hangcheck.c +++ b/drivers/gpu/drm/i915/selftests/intel_hangcheck.c @@ -33,6 +33,7 @@ struct hang { struct drm_i915_private *i915; struct drm_i915_gem_object *hws; struct drm_i915_gem_object *obj; + struct i915_gem_context *ctx; u32 *seqno; u32 *batch; }; @@ -45,9 +46,15 @@ static int hang_init(struct hang *h, struct drm_i915_private *i915) memset(h, 0, sizeof(*h)); h->i915 = i915; + h->ctx = kernel_context(i915); + if (IS_ERR(h->ctx)) + return PTR_ERR(h->ctx); + h->hws = i915_gem_object_create_internal(i915, PAGE_SIZE); - if (IS_ERR(h->hws)) - return PTR_ERR(h->hws); + if (IS_ERR(h->hws)) { + err = PTR_ERR(h->hws); + goto err_ctx; + } h->obj = i915_gem_object_create_internal(i915, PAGE_SIZE); if (IS_ERR(h->obj)) { @@ -79,6 +86,8 @@ err_obj: i915_gem_object_put(h->obj); err_hws: i915_gem_object_put(h->hws); +err_ctx: + kernel_context_close(h->ctx); return err; } @@ -196,9 +205,7 @@ unpin_vma: } static struct drm_i915_gem_request * -hang_create_request(struct hang *h, - struct intel_engine_cs *engine, - struct i915_gem_context *ctx) +hang_create_request(struct hang *h, struct intel_engine_cs *engine) { struct drm_i915_gem_request *rq; int err; @@ -225,7 +232,7 @@ hang_create_request(struct hang *h, h->batch = vaddr; } - rq = i915_gem_request_alloc(engine, ctx); + rq = i915_gem_request_alloc(engine, h->ctx); if (IS_ERR(rq)) return rq; @@ -244,6 +251,58 @@ static u32 hws_seqno(const struct hang *h, return READ_ONCE(h->seqno[rq->fence.context % (PAGE_SIZE/sizeof(u32))]); } +struct wedge_me { + struct delayed_work work; + struct drm_i915_private *i915; + const void *symbol; +}; + +static void wedge_me(struct work_struct *work) +{ + struct wedge_me *w = container_of(work, typeof(*w), work.work); + + pr_err("%pS timed out, cancelling all further testing.\n", + w->symbol); + i915_gem_set_wedged(w->i915); +} + +static void __init_wedge(struct wedge_me *w, + struct drm_i915_private *i915, + long timeout, + const void *symbol) +{ + w->i915 = i915; + w->symbol = symbol; + + INIT_DELAYED_WORK_ONSTACK(&w->work, wedge_me); + schedule_delayed_work(&w->work, timeout); +} + +static void __fini_wedge(struct wedge_me *w) +{ + cancel_delayed_work_sync(&w->work); + destroy_delayed_work_on_stack(&w->work); + w->i915 = NULL; +} + +#define wedge_on_timeout(W, DEV, TIMEOUT) \ + for (__init_wedge((W), (DEV), (TIMEOUT), __builtin_return_address(0)); \ + (W)->i915; \ + __fini_wedge((W))) + +static noinline int +flush_test(struct drm_i915_private *i915, unsigned int flags) +{ + struct wedge_me w; + + cond_resched(); + + wedge_on_timeout(&w, i915, HZ) + i915_gem_wait_for_idle(i915, flags); + + return i915_terminally_wedged(&i915->gpu_error) ? -EIO : 0; +} + static void hang_fini(struct hang *h) { *h->batch = MI_BATCH_BUFFER_END; @@ -255,7 +314,9 @@ static void hang_fini(struct hang *h) i915_gem_object_unpin_map(h->hws); i915_gem_object_put(h->hws); - i915_gem_wait_for_idle(h->i915, I915_WAIT_LOCKED); + kernel_context_close(h->ctx); + + flush_test(h->i915, I915_WAIT_LOCKED); } static bool wait_for_hang(struct hang *h, struct drm_i915_gem_request *rq) @@ -290,7 +351,7 @@ static int igt_hang_sanitycheck(void *arg) if (!intel_engine_can_store_dword(engine)) continue; - rq = hang_create_request(&h, engine, i915->kernel_context); + rq = hang_create_request(&h, engine); if (IS_ERR(rq)) { err = PTR_ERR(rq); pr_err("Failed to create request for %s, err=%d\n", @@ -427,8 +488,7 @@ static int __igt_reset_engine(struct drm_i915_private *i915, bool active) struct drm_i915_gem_request *rq; mutex_lock(&i915->drm.struct_mutex); - rq = hang_create_request(&h, engine, - i915->kernel_context); + rq = hang_create_request(&h, engine); if (IS_ERR(rq)) { err = PTR_ERR(rq); mutex_unlock(&i915->drm.struct_mutex); @@ -487,7 +547,9 @@ static int __igt_reset_engine(struct drm_i915_private *i915, bool active) if (err) break; - cond_resched(); + err = flush_test(i915, 0); + if (err) + break; } if (i915_terminally_wedged(&i915->gpu_error)) @@ -633,8 +695,7 @@ static int __igt_reset_engine_others(struct drm_i915_private *i915, struct drm_i915_gem_request *rq; mutex_lock(&i915->drm.struct_mutex); - rq = hang_create_request(&h, engine, - i915->kernel_context); + rq = hang_create_request(&h, engine); if (IS_ERR(rq)) { err = PTR_ERR(rq); mutex_unlock(&i915->drm.struct_mutex); @@ -726,7 +787,9 @@ unwind: if (err) break; - cond_resched(); + err = flush_test(i915, 0); + if (err) + break; } if (i915_terminally_wedged(&i915->gpu_error)) @@ -787,7 +850,7 @@ static int igt_wait_reset(void *arg) if (err) goto unlock; - rq = hang_create_request(&h, i915->engine[RCS], i915->kernel_context); + rq = hang_create_request(&h, i915->engine[RCS]); if (IS_ERR(rq)) { err = PTR_ERR(rq); goto fini; @@ -866,7 +929,7 @@ static int igt_reset_queue(void *arg) if (!intel_engine_can_store_dword(engine)) continue; - prev = hang_create_request(&h, engine, i915->kernel_context); + prev = hang_create_request(&h, engine); if (IS_ERR(prev)) { err = PTR_ERR(prev); goto fini; @@ -880,9 +943,7 @@ static int igt_reset_queue(void *arg) struct drm_i915_gem_request *rq; unsigned int reset_count; - rq = hang_create_request(&h, - engine, - i915->kernel_context); + rq = hang_create_request(&h, engine); if (IS_ERR(rq)) { err = PTR_ERR(rq); goto fini; @@ -952,6 +1013,10 @@ static int igt_reset_queue(void *arg) i915_gem_chipset_flush(i915); i915_gem_request_put(prev); + + err = flush_test(i915, I915_WAIT_LOCKED); + if (err) + break; } fini: @@ -989,7 +1054,7 @@ static int igt_handle_error(void *arg) if (err) goto err_unlock; - rq = hang_create_request(&h, engine, i915->kernel_context); + rq = hang_create_request(&h, engine); if (IS_ERR(rq)) { err = PTR_ERR(rq); goto err_fini; diff --git a/drivers/gpu/drm/i915/selftests/mock_context.c b/drivers/gpu/drm/i915/selftests/mock_context.c index bbf80d42e793..501becc47c0c 100644 --- a/drivers/gpu/drm/i915/selftests/mock_context.c +++ b/drivers/gpu/drm/i915/selftests/mock_context.c @@ -92,3 +92,14 @@ live_context(struct drm_i915_private *i915, struct drm_file *file) return i915_gem_create_context(i915, file->driver_priv); } + +struct i915_gem_context * +kernel_context(struct drm_i915_private *i915) +{ + return i915_gem_context_create_kernel(i915, I915_PRIORITY_NORMAL); +} + +void kernel_context_close(struct i915_gem_context *ctx) +{ + context_close(ctx); +} diff --git a/drivers/gpu/drm/i915/selftests/mock_context.h b/drivers/gpu/drm/i915/selftests/mock_context.h index 2f432c03d413..29b9d60a158b 100644 --- a/drivers/gpu/drm/i915/selftests/mock_context.h +++ b/drivers/gpu/drm/i915/selftests/mock_context.h @@ -36,4 +36,7 @@ void mock_context_close(struct i915_gem_context *ctx); struct i915_gem_context * live_context(struct drm_i915_private *i915, struct drm_file *file); +struct i915_gem_context *kernel_context(struct drm_i915_private *i915); +void kernel_context_close(struct i915_gem_context *ctx); + #endif /* !__MOCK_CONTEXT_H */ |
